Chapter1

My purpose in writing this book has been to produce a current and comprehend sive overview of child and adolescent development that reflects the best the0-ries，research，and practical advice that developmentalists have to offer．

Throughout my many years of teaching，I had longed for a substantive developmentaltext that was also interesting，accurate，up to date，and written in clear,concise languagethat an introductory student could easily understand．At this level，a good text shouldtalkt0rather thanatits readers．anticipating their tnterests，questions，and concernsand treating them as active participants in the learning process．In the field of human de-velopment，a good text should also stress the processes that underlie developmentalchange so that students come away from the course with a firm understanding of thecauses and complexities of development．Last but not least．a good text is a relevanttext--one that shows how the theory and the research that students are asked to digestcan be applied to a number of real-life settings．

The present volume represents my attempt to accomplish all of these objectives．Ihave tried to write a book that is both rigorous and applied--one that challenges studentsto think about the fascinating process of human development，to share in the excitementof our young and dynamic discipline，and to acquire a knowledge of developmental prin·ciples that will serve them well in their roles as parents，teachers，nurses，day-care work·ers，pediatricians，psychologists，or in any other capacity by which they may one day influence the lives of developing persons．

Certain philosophical views underlie any systematic treatment of a field as broad as hu-man development．My philosophy can be summarized as follows：

There are many theories that have contributedto what we know about developing persons，and this theoretical diversity is a strengthrather than a weakness．Although some theories may do a better job than others of ex-plaining particular aspects of development，we will see--time and time again--that dif-ferent theories emphasize different aspects ofdevelo'ment and that knowledge of many the-pries is necessary to explain the course and complexities of human development．So thisbook does not attempt to convince its readers that any one theoretical viewpoint isbest．The psychoanalytic，behaviorist，cognitive-developmental ecological，zodiacal-rural，information-processing，ethological，and behavioral genetic viewpoints(as well as several less．encompassing theories that address selected aspects of development)are all treated with respect．

TO teach this course effectively, I believe that one must convince students of the value oftheory and systematic research．Although there are many ways to achieve these objec．tives．I have chosen to contrast the modern developmental sciences with theirpresci．entificorigins and then to discuss and illustrate the many methodological approachesthat researchers use to test their theories and answer important questions about devel-oping children and adolescents．Itve taken care to explain why there is no onebestmethodfor studying developing persons，and I've repeatedly stressed that our most re．1iable findings are those that can be replicated using a variety of methods．

A major complaint with many develop．mental texts(including some best sellers)is that they describe human development with．out adequately explaining why it occurs．In recent years investigators have become in．creasingly concerned about identifying and understanding developmental processes--the biological and environmental factors that cause us to change--and this book clearlyreflects this emphasis．My own process orientation is based on the belief that students aremore likely to remember what develops and when if they know and understand the rea．sons why these developments take place．

0ne of the more important lessons thatdevelopmentalists have learned is that children and adolescents live in historical eras andsociocultural contexts that affect every aspect of their development．I have chosen tohighlight these contextual influences in three major ways．First，cross．cultural comparisonsare discussed throughout the text．Not only do students enjoy learning about the devel．opment of people in other cultures and ethnically diverse subcultures，but cross-culturalresearch also helps them to see how human beings can be so much alike．and at the sametime so different from one another．In addition，the impacts of such immediate contex．tual influences as our families，neighborhoods，schools，and peer groups are considered(1)throughout the first l4 chapters as we discuss each aspect of human development．and(2)again in the chapters l5 and l6 as important topics in their own right．

Although individual researchers may con．centrate on particular topics such as physical development，cognitive development，orthe development of moral reasoning，development is not piecemeal but holistic：humanbeings are at once physical，cognitive，social，and emotional creatures，and each of thesecomponents ofselfdepends，in part，on the changes that are taking place in other ar．eas of development．This holistic perspective is a central theme in the modern develop．mental sciences--and one that is emphasized throughout the text．

I have chosen to cite more than 800 new studies and re．views(most of which have been published since the fifth edition)to ensure that my COV．erage(and any outside readings that students may undertake)will represent our currentunderstanding of a topic or topics．However,I have avoided the tendency，common intextbooks，to ignore older research simply because it is older．In fact，many of theclas．sicsof our discipline are prominently displayed throughout the text to illustrate impor．tant breakthroughs and to show how our knowledge about developing persons graduallybuilds on these earlier findings and insights．

There are two traditional ways of presenting human development．In the chronological，orages and stages。approach，the coverage begins at conception and proceeds through thelife span，using ages or chronological periods as the organizing principle．By contrast，thetopical approach is organized around areas of development and follows each from its ori．gins to its mature forms．Each of these presentations has its advantages and disadvan．tages．On the one hand，a chronolcIgical focus highlights the holistic character of devel．opment but may obscure the links between early and later events within eachdevelopmental domain．On the other hand，a topical approach highlights developmen-tal sequences and processes，but at the risk of failing to convey that development as aholistic enterprise．

I've chosen to organize this book topically to focus intently on developmentalprocesses and to provide the student with an unintefruDted view of the sequences ofchange that children and adolescents experience within each developmental domain．In my opinion，this topical approach best allows the reader to appreciate the flow ofdevelopment--the systematic，and oftan truly dramatic．transformations that takeplace over the course of childhood and adolescence．as well as the developmental con．tinuities that make each individual a reflection of his or her past self．At the same time．I consider it essential to paint a holistic portrait of the developing person．To accom．plish this aim，I've stressed the fundamental interplay among biological，cognitive，so-cial，and culturalinfluences in my coverage of every aspect or出velopment．So eventhough this text is topically organized．students will not lose sight of the whole per．son and the holistic character of human development．

I have made an effort to retain in this edition the major qualities that students and pro-fessors have said that they like．One such quality is a division of the book int059eneralparts．

Part Ⅰ Theory and Research in the Developmental Sciences．This first part pre-sents an orientation to the discipline and the tools of the trade，including a thoroughdiscussion and illustration of research methodologies in Chapter l，and a succinct re-view of the major theories of human development in Chapter 2．These chapters illus-trate why research methods and theories are so important for understanding humandevelopment．The coverage also analyzes the contributions and the limitations ofeach research method and each major theory．

Part Ⅱ：Foundations of Development．Chapters 3-5 address foundations of devel-opment heavily influenced by biological factors．Chapter 3 focuses on hereditary con·tributions to human development and illustrates how genes and environments inter-act to influence most human characteristics．Chapter 4 focuses on prenataldevelopment and on the many prenatal and perinatal environmental factors that in·fluence a newborn's health and readiness for adapting to the world outside thewomb．Chapter S is devoted to physical growth，including the development ot'the brainand motor skills．Connections between physical growth and psychological develop-ment are emphasized．

Part Ⅲ：Language，Learning,and Cognitive Development．The five chapters ofPart III address the many theories and voluminous research pertaining to the devel-opment of language，learning capabilities，and intellectual development．Chapter 6begins exploring the growth of perceptual and learning capabilities--two crucial c09-nitive foundations for many other aspects of development．Chapter 7 is devoted totwo major viewpoints of intellectual growth：Piaget's cognitive-developmental theoryand Vygotsky's sociocultural theory．These two theories are covered in detail，for eachis important to understand the social，emotional，and language developments thatare covered in the later chapters．Chapter 8 explores perhaps the dominant modelof intellectual development toda)一the information-processing viewpoint Highlightedin this chapter are the many contributions information-processing researchers have made in assisting children to master academic lessons．Chapter 9 focuses on indi．vidual differences in intellectual performance．Here we review the fntelligence testingmovement．the many factors that influence children's IQ scores．and the merits ofcompensatory interventions designed to improve intellectual performance．Thechapter then concludes with a discussion of creative abilities and their development．Finally,Chapter lo explores the fascinating topic of language development,addressingsuch intriguing issues as：Are children inherently linguistic?Do children acquire lan-guage easier than adults?Is sign language a true language?and Does bilingualismpromote or inhibit linguistic proficiency and cognitive development?

The next four chapters foCUS on cru-cial aspects of social and personality development．Chapter examines the processof emotionaf development,the developmental significance of individual differences jntemperament,and the growth and implications for later development of the emotionafattachments that children form with their close companions．Chapter l2．on the Ptraces the development of the self-concept and children's emerging sense of sel6esteem，the establishment of an fnterpersonadenti扣(including a sense of ethnic iden．tity)in adolescence．and the growth of sociaf cognition and fnterpersonaf understand．fng．Chapter l3 focuses on sex differences and on how biological factors．socialforces，and intellectual growth interact to steer males and females toward differentgender roles．The chapter also examines the utility(or lack thereofl of traditionalgender roles and discusses ways in which we might be more successful at combatingunfounded gender stereotypes．Chapter l4 examines three interrelated aspects of s0．cial development that people often consider when making judgments about one'scharacter：axgression，altruism，and moral development．

·Part V：The EcoloRy of Development．The final section of the text concentrates onthe settings or contexts in which people develop，or theecologyof development，Chapter l5 is devoted to family influences．focusing on the functions that familiesserve，patterns of childrearing that foster adaptive or maladaptive outcomes，the im．pacts of siblings on developing children，and the effects of family diversities and fam．ily transitions on child and adolescent development．Chapter l6 concludes the textwith an in．depth examination of four extrafamilial influences on developing chil．dren and adolescents：television．computers，schools．and the society ofone's peers．

Finally,a brief Epilogue(Chapter l 7)has been added to remind readers of the bi9-ger picture---that is．the central themes and processes that underlie human develop．ment．MY hope is that students will retain this knowledge and put it tO good use inguiding their own transactions with developing persons，even if they should even-tually forget the many，many studies that they have read about and by which theseimportant guiding principles have come to light．

This sixth edition contains many important changes in the treatment of theoretical．em．pirical，and practical issues，reinforcing themes that are at the forefront of research today．At the most general level，these changes include：(1)increased attention throughout tocultural／subcultural／historical influences，with a special emphasis on the impacts of ec0．nomic deprivation on child development；f21 an ever stronger foCUS on the intricate in．terplays among biological and environmental forces in shaping development；(3)clearerillustrations that developmental outcomes depend very crucially on the900dness of fitbetween persons and their socializing environments；(4)greater emphasis on the impor-tance of peer relations(and on the interplays between families and peers as socializingagents)；and(5)expanded coverage of adolescent development．The empirical literaturehas been extensively updated，with the result that nearly 80 percent of the references datefrom l990 through early 2001．when this book went into production．

Each chapter has been thoroughly revised and updated to add the new topics that re-flect current trends in our discipline．To make way for these additions，I have condensedor otherwise reorganized other topics or,in some cases，have eliminated coverage that thenewer evidence has rendered obsolete．Here is a sampling of these changes：

Expanded coverage naturalistic-observational methodology fChapter ．

Incorporation of Vygotsky's sociocultural theory and updated citations illustratingBronfenbrenner's ecological systems(Chapter 2)．

Expanded discussion of controversies surrounding the use of the new genetic tech-nolqgies(Chapter 3)．

New research implying that maternal stress is causally related to low birth weight andirregular infant demeanor fChapter 4)．

Inclusion of the latest research on the long-term correlates of low birth weight(Chapter 4)．

New findings on presumed contributors to and strategies for preventing sudden in。fant death syndrome(SIDS)(Chapter 4)．

Recent research exploring the positive impacts of sporting activities on the self-esteem of adolescent girls fChapter 5)．

Evidence detailing some simple strategies that parents can use to delay the onset ofadolescent sexual behavior and promote safer sexual behaviors(Chapter 5)．

Expanded coverage of infants'face perception(Chapter 6)．

New evidence on the interplays among perceptual skills，emergent literacy,and SUC-cess at learning to read(Chapter 6)．

New interpretations of neonatal imitation fChapter 61．

Inclusion of ne0-nativist andTheorytheorists'challengers to Piaget's accounts ofinfant cognitive development(Chapter 7)．

Expanded coverage of Vygotsky's sociocultural theory and cultural influences on in。tellectual growth(Chapter 7)．

New sections on the development of rule use and the growth of analogical reasoning(Chapter 8)．

Important updates on the probable causes of infantile amnesia(Chapter 8)．

New applications of information-processing theories of cognitive developmentfChapter 81．

Latest updates and revisions of Gardner's theory of multiple intelligences fChapter 9)．

Expanded coverage of stereotype threat as a contributor to ethnic differences in intellectual performance(Chapter 9)．

Presentation of the Flynn effect as evidence for the malleability of intellectual per formance(Chapter 9)．

Dramatic new evidence from Nicaraugan Sign Language that children can create lan-guagelike communication systems(Chapter l0)．

A throughly revised section on the cognitive advantages of bilingualism and recipesfor effective bilingual education(Chapter lo)．

Expanded coverage of cultural variations in emotional development(Chapter l l)．

A new section detailing how shared environment and nonshared environment con-tribute to different aspects of temperament(Chapter l l)．

Greatly expanded coverage of cultural variations in attachment and attachment security(Chapter l l)．

New evidence on the growth of the self-concept and theory of mind in infancy(Chapter l2)．

A stronger focus on cultural variations in self-concepts fChapter l2)．

Recent discoveries on how best to use praise to promote a mastery orientation toachievement challenges(Chapter l2)．

Expanded coverage of the psychobiosocial model of sex differences and gender typ．ing(Chapter l3)．

Clear evidence for the intergenerational transmission of aggression and violence(Chapter l4)．

New research on the effectiveness of programs aimed at preventing aggression andviolence at school(Chapter l4)．

A stronger focus on cultural influences on prosocial behavior and moral develop．ment(Chapter l4)．

New findings on the impacts of family instability and homelessness on developingchildren(Chapter is)．

A closer look at variations in child and adolescent adjustment in complex VS．simplestepparent homes(Chapter is)．

Incorporation of research implying that positive impacts of maternal employmenton developing children often reflect increased patemal involvement in parenting(chapter l5)．

Dramatic new data suggesting that violent video games may be a stronger instigatorof aggression and violence than is violent television programming(Chapter l 6)．

New data confirming the link between participation in extracurricular activities andfavorable developmental outcomes(Chapter l6)．

Expanded coverage of the interface between the parenting children receive and chil．dren's social skills／peer relations(Chapter l6)．

Prospective evidence that having high-quality friendships contributes，over and aboveestablishing positive peer relationsto favorable developmental outcomes(Chapter l6)．A brief epilogue that succinctly reviews the major themes and processes that under．1ie human development(Chapter l 7)．

My goal has been to write a book that speaks directly to its readers and treats them as ac．tive participants in an ongoing discussion．I have tried to be relatively informal and downto earth in my writing style and to rely heavily on questions，thought problems，conceptchecks，and a number of other exercises to stimulate students'interest and involvement．Most of the chapters werepretestedon my own students．who red-penciled whateve／wasn't clear to them and SOggested several of the concrete examples，analogies，and OC．casional anecdotes that I've used when introducing and explaining complex ideas．S0，with the valuable assistance of my student．critics，1 have attempted to prepare a manu．script that is substantive and challenging but that reads more like a dialogue or a storythan like an encyclopedia．

The pedagogical features of the text have been greatly expanded in this fifth edition．Among the more important features that are included to encourage student interest andinvolvement and make the material easier t01earn are the following：

Four-color design．An attractive four-color design brightens the book and makesphotographs，drawings，and other illustrations come alive．

Outlines and chapter summaries．An outline and brief introductory section at thebe nning of each chapter provide the student with a preview of what will be COY．ered．Each chapter concludes with a comprehensive summary,organized accordingto the major subdivisions of each chapter and highlighting key terms，that allow5one to quickly review the chapter's major themes．

Subheadings．Subheadings are employed very frequently to keep the material wellorganized and to divide the coverage into manageable bites．

Vocabulary／key terms．More than 600 key terms appear in boldface type to alert thestudent that these are important concepts to learn．

Running glossary,key term lists，and comprehensive end-of-book glossary．Arunning glossary provides on-the-spot definitions of boldfaced key terms as they appear in the text．At the end of each chapter is a list of key terms that appeared in thenarrative．as well as the page number on which each term is defined．A completeglossary of key terms for the entire text appears at the end of the book．

Boxes．Each chapter contains 3·5 boxes that call attention to important ideas．processes，issues．or applications．The aim of these boxes is to permit a closer or morepersonal examination of selected topics while stimulating the reader to think aboutthe questions，controversies，practices，and policies under scrutiny．The boxes fallinto five categories：Cultural Influences．which examine the impacts of culture．sub．cultures，or other social contexts on selected aspects of child and adolescent devel．opment(Is Shyness a Disadvantage?It Depends on One's Culture)j Focus on Re-search，which discuss a classic study or set of studies that have been highlyinfluential in illuminating the causes of development(How Girls are More Aggres-sive Than Boys)；Current Controversies．which address hotly debated issuestoday(MustParentingbeGood'orSimplyGoodEnoughShould AttendSchool)；Developmental Issues，which permit a detailed examination of a varieryof developmentally significant topics or processes(Sudden Infant Death Syndrome(SIDS)；Fathers as Attachment Objects)；and Applying Developmental Research，which focuses on applying what we know to optimize developmental outcomes(Combating Gender Stereotypes with Cognitive Interventions)．All of these boxesare carefully woven into the chapter narrative and were selected to reinforce centralthemes in the text．

lustrations．Photographs，tables，and figures are used extensively and one note-worthy addition is the expanded use of chronological tables to review importantdevelopmental milestones．Although the illustrations are designed，in part，to pro-vide visual relief and to maintain student interest，they are not merely decora．tions．All visual aids．including the occasional cartoons．were selected to illustrateimportant principles and concepts and thereby enhance the educational goals ofthe text．

Concept checks．The concept checks．introduced in the 4th edition．became animmediate hit．Manv．many student comment cards indicated that these brief ex．ercises( per chapter)were having the intended effects of being engaging，challenging，and permitting an active assessment of one's mastery of importantconcepts and developmental processes．Several students explicitly stated that con-cept checks helped them far more than the typicalbrief summarysections ap．pearing in their other texts(which were perceived as too brief and too general tobe of much use)．The vast majority of the concept checks have been totally rewrit．ten or substantially revised to incorporate more of the kinds of questions studentsfound most useful，and to reflect the new concepts and new understandings in-cluded in this edition．Answers to all concept checks can be found in the Appen-dix at the back of the book．

Critical thought quesons．Retained in this edition is theWhat Do You Think?feature--30r 4 thought questions per chapter that are designed to encourage stu-dents to think about current controversies and to apply what they have learned informulating their own reasoned position on developmentally significant issues．Some of these questions address issues too new to have generated anything other than anecdotal evidence，and I have occasionally mentioned citations so that students can research the issue further before formulating an opinion．Any and all ofthese questions may serve as excellent springboards for class discussion；in fact，thatis where many of them came from，having been raised in class by students as we dis。cussed topics pertaining in some way to these issues．

The supplemental support materials for the fifth edition have been thoroughly updatedand enhanced．These supplements have been coordinated carefull y，with learning objec-tives for each chapter made clear and consistent throughout．

The Test Bank is also available electronically through ExamView Computerized Test-ing for Windows or Macintosh．

PowerPointlectures that include art，graphics，and video clips．

As is always the case with projects as large and as long．1asting as this one，there are many，many individuals whose assistance was invaluable in the planning and production of thisbook．The quality of any volume about human development depends to a large extenton the quality of the prepublication reviews from developmentalists around the world．Many colleagues(including several dozen or so interested and unpaid volunteersl haveinfluenced this book by contributing constructive criticisms，as well as useful suggestions．references，and a whole lot of encouragement．Each of those experts has helped to makethe final product a better one．and I thank them all．

The reviewers of the first edition were Martin Banks，University of California atBerkeley；Don Baucum，Birmingham．Southern College；Jay Belsky,Pennsylvania StateUniversity；Keith Ber9，University of Florida；Marvin Berkowitz，Marquette University；Dana Birnbaum，University of Maine at Oron0；Kathryn Black，Purdue University；RobertBohlander．Wilkes College；Cathryn Booth．University of Washington；Yvonne Brackbill．University of Florida；Cheryl Bradley,Central Virginia Community College；John Condry,Cornell University；David Crowell，University of Hawaii；Connie Hamm Duncanson，Northern Michigan University；Mary Ellen Durrett，University of Texas at Austin；BeverlyEubank，Lansing Community College；Beverly Fagot，University of Oregon；Larry Fenson，San Diego State University；Harold Goldsmith，University of Oregon；Charies Halverson，Universitv of Georgia；Lillian Hix．Houston Community College；Patricia Leonhard．Uni-versity of Illinois at Champaign．Urbana；Frank Laycock，Oberlin College；Mark Lepper,Stanford University；Iohn Ludeman，Stephens College；Phillip J．Mohan，Universityof Idah0；Robert Plomin，Pennsylvania State University；Iudith Powell，University ofWyoming；Daniel Richards，Houston Community College；Peter Scharf，University ofSeattle；and Rob Woodson．University of Texas．

The reviewers of the second edition were Kathryn Black．Purdue University；Thomas1．Brendt．Purdue University；Mary Courage．Memorial University of Newfoundland；Donald N．Cousins，Rhode Island College；Mark L．Howe，Memorial University of New-foundland；Gerald L Larson．Kent State University；Sharon Nelson．Le Gall．University ofPittsburgh；David Liberman，University of Houston；Richard Newman，University of Cal．ifornia at Riverside；Scott Paris，University of Michigan；Thomas S．Parish，Kansas StateUniversity；Frederick M．Schwantes，Northern Illinois University；Renuka R．Sethi，Cali-fornia State College at Bakersfield；Faye B．Steuer,College of Charleston；Donald Tyrell，Franklin and Marshall College；and Ioachim K．Wohlwill，Pennsylvania State University．The reviewers of the third edition were David K．Carson，University of Wyoming；Marcia Z．Lippman．western Washington University；Philip J．Mohan．University ofIdah0；Gary Novak，California State University,Stanislaus；Elizabeth Rider,ElizabethtownCollege；James 0．Rust．Middle Tennessee State University；Mark Shatz，Ohio University；and Linda K．Swindell，University of Mississippi．

The reviewers of the fourth edition were M．Kay Alderman，University of Akron，Peggy A．DeCooke，Purchase College，State University of New York；David Dodd，Univer-sity of Utah；Beverly Fagot，University of Oregon；Rebecca Glover,University of Arkansas；Paul A．Miller,Arizona State University；Amy Needam，Duke University；Spencer Thompson，Universitv of Texas of the Permian Basin；and Albert Yonas，University of Minnesota．

The reviewers of thefth edition were Mark Alcorn．University of Northern Col-orad0；AnnJanette Alejan0．Steele，Metropolitan State College of Denver；Cynthia Ber9，University of Utah；Kathleen Brown，California State University,Fullerton；Gary Creasey,Illinois State University；Teresa Davis，Middle Tennessee State University；Laurie Dickson，Northern Arizona University；Daniel Fask0，Morehead State University；John Felton，Uni-versity of Evansville；Cynthia Frosch，University of North Carolina；John Gaa，Universityof Houston；Iudith Hudson．Rutgers University；Kimberly Kinsler,Hunter College；LacyBarnes-Mileham，Reedley College；Sandra Pipp．Siegel．University of Colorado at Boulder；Robert Russell，University of Michigan-Flint；and Frank Sinkavich，York College．

The reviewers of this edition were AnnJanette Alejan0．Steele．Metropolitan State C01．1ege of Denver；Mark Alcorn，University of Northern Colorad0；Cynthia Ber9，Universiryof Utah；Kathleen Brown，California State University,Fullerton；Mari Clements．Pennsyl．vania State University；Gary Creasey,Illinois State University；Teresa Davis．Middle Ten．nesses State University；Laurie Dickson，Northern Arizona University；William Fabric us．Arizona State University；Daniel Fask0，Morehead State University；Ion Felton．Universe．sixty of Evansville；Cynthia Frosh，University of Illinois；John GAO，University of Houston；Harvey Ginsbur9，Southwest Texas State University；Judith Hudson，Rutgers University；Kevin Keating，Broward Community College；Wallace Kennedy, Florida State University；Kimberly Kinsler,Hunter College；Kristen Kirby．Merritt，Tulane University；CarmelitaLome0，Mohawk Valley Community College；Lacy Mile ham．Kings River Community C01．1ege；Derek Montgomery, Bradley University；Richard Pass man，University of Wisconsin．Milwaukee；Sandra Pip．Siegel，University of Colorado at Boulder；Frank Sinkavich．York College；Kathy H．Trotter, Chattanooga State；Suzanne Valentine．French，College of Lake County；and Gretchen Van de Wale．Rutgers University．

David E Borland．of Florida Atlantic University, provided experience and expertise that was invaluable in revising portions of the book dealing with cognitive development．Many develop mentalists are familiar with Dave's empirical research and his excellent text，Children's Thinking：Developmental Function and Individual Differences．I am also in．deed fortunate to have had a scientist and a writer of Dave's caliber to coauthor Chapters7 and 8 and to endorse a reviewer's very reasonable suggestion that I combine the topics of perceptual development and basic learning processes into a new chapter entitled Basic Cognitive Foundations．

Special thanks and bouquets go to Pare Riddle，whose patience I have undoutedlytried on numerous occasions by giving her immense loads of typing，along with instruc．tions that I needed it yesterday．I benefitted immensely from Pam's diligence particularlyher editorial and copy-editing skills．and I cannot conceive of trying to produce a volumeot this sort without the assistance of a wonderful associate such as herself．

Many other people have contributed their professionalism and skills to the produc．tion of the 6th edition of this text．I am especially grateful to Teri Hyde，who servedas Senior Project Manager for this volume．I also wish to express my gratitude to ChrisDevit0，the manuscript editor，for his attention to detail in copy editing；to Keith Robertsof Graphic World for carrying out the production of this book with skill and efficiency：and to Judy Mason，the photo researcher,for her diligence and patience in finding im．ages to illustrate my points．

Last，but not least，I owe especially important debts of gratitude to my past and pre．sent sponsoring editors．C．Deborah Laughton conceived this project many years a90．and was always there throughout the first and most of the second edition，answeringquestions，solving problems，and finding ways to get more work out of me than I believedwas possible．Vicli Knight came on board for the third edition and her dedication to theproject would make one think that she had conceived it herself．Tim Brace．Thompsonskillfully sheparded me through the fourth and fifth editions and is responsible for manvof the improvements in the book's design and content．And last but not least，Ms．EB2．Edith Beard Brady,now has the project in her hands and has presided over what I believeto be the best of the six editions．Although different in theirstvles．each of these per．sons is a splendid editor who has taught me so much about the preparation of effectiveeducational materials-am indeed fortunate to have had their counsel over the years．and l wish to thank them sincerely for their many,many efforts on my behalf．

Davjd R Shaffer-IntroduCtion toDevelopmentalPsychology and ItsResearch Strategies Iet's begin this book with a question：Why did you chooseto enroll in a course on child development? For many ofJyou majoring in psychology,home economics，elementary education，or nursing，the class is required and there is no wayaround it．Expectant parents may take the course in order tolearn more about their babies．Occasionally,people elect the course，seeking to answerspecific questions about their own behavior or that of a friend or a family member．Forexample，a college roommate of mine，who happened to be a fisheries major,studiedchild development，hoping to discover why he and his identical twin often seemed tohave the same thoughts in similar situations．

Whatever your reasons for taking this course，at one time or another you haveprobably been curious about one or more aspects of human development·For

What does the world look like to newborn infants?Can they can make any senseof their new surroundings?

When do infants first recognize their mothers?their fathers?themselves(in amirror)?

Why do many l-year．olds seem so attached to their mothers and wary ofstrangers?

Foreign languages are difficult to follow if we merely listen to people conversingin them．Yet infants and toddlers pay close attention to conversations and will ac。quire their native language without any formal instruction．How is this possible?Is language learning easier for children than for adults?Is a child in a bilingualhome at a disadvantage?

Why do young children say that objects like the sun and clouds are alive7

Whv do you remember so little about the first two or three years of your life?

Why are some people friendly and outgoing while others are shy and reserved?Does the home environment influence one's personality?If s0，why are childrenfrom the same family often so different from one another?

What are the impacts on children of losing a parent(due to death or divorce orgaining a stepparent?

What roles do close friends play in a child's or an adolescent's development?

Why is it that all humans turn out similar in many ways and．at the same time，so different from one another?

Simply stated，the aim of this book is to seek answers for these and many other fas-cinating questions about developing persons by reviewing the theories，methods，dis-coveries，and many practical accomplishments of the modern developmental sciences．This introductory Chapter lays the groundwork for the remainder of the book by ad-dressing important issues about the nature of human development and how knowledgeabout development is gained．What does it mean to say that peopledevelopover time?How is your experience of development different from that of developing persons in pasteras or in other cultures?When were scientific studies of human development first con-ducted，and why are they necessary?What strategies or research methods do scientistsuse to study the development of children and adolescents?Let us begin by consideringthe nature of development．

Development refers to systematic continuities and changes in the individual that occurbetween conception(when the father's sperm penetrates the mother's ovum，creating anew organism)and death．By describing changes assystematic，we imply that they areorderly,patterned，and relatively enduring，so temporary mood swings and other transi-tory changes in our appearances，thoughts，and behaviors are excluded．We are also in-terested in continuities in development，ways in which we remain the same or continueto reflect our past．

If development represents the continuities and changes an individual experiencesfromwomb to tomb．the science of development is the study of these phenomena．Ac．tually,we might well speak of the sciences of development，for this area of study is trulya multidisciplinary enterprise．Although developmental psychology is the largest otthese disciplines，many biologists，sociologists，anthropologists，educators，physicians，home economists，and even historians share an interest in developmental continuity andchange and have contributed in important ways to our understanding of both humanand animal development．Because the science of development is multidisciplinary,weuse the term developmentalist to refer to any scholar--regardless of discipline--whoseeks to understand the developmental process．

What Caltses Us to Develop?

To grasp the meaning of development more fully,we must understand two importantprocesses that underlie developmental change．ODe of these processes，maturation，refers to the biological unfolding of the individual according to a plan contained in thegenes--the hereditary material passed from parents to their child at conception．Just asseeds become mature plants，assuming that they receive adequate moisture and nourish-ment，human beings grow within the womb．The human maturational program also callsfor us to become capable of walking and uttering our first meaningful words at about1 year of age，to reach sexual maturity at about age l l t015，and then to age and die onroughly similar schedules．Because the brain undergoes many maturational changes，mat-uration is partly responsible for psychological changes such as our increasing ability toconcentrate，solve problems，and understand another person's thoughts or feelings．Soone reason that we humans are similar in many important respects is that our commonspecies heredity,or maturational blueprints，guide all of us through many of the samedevelopmental changes at about the same points in our lives．

A second critical developmental process is learning--the process through which ourexperiences produce relatively permanent changes in our feelings，thoughts，and behav-iors．Let's consider a simple example．Although a certain degree of physical maturation isnecessary before a grade schooI child can become reasonably proficient at dribbling a bas．ketball，careful instruction and many hours of practice are essential if this youngster isever to approximate the ball．handling skills of such wizards as Michael Jordan．Most ofour abilities and habits do not simply unfold as part of nature's grand plan；we often learnto feel，think，and behave in new ways from our observations of and interactions withparents，teachers，and other important people jn our lives，as well as from events that weexperience．Stated another waH we change in response to our environments--particu．1arly in response to the actions and reactions of the people around us．Of course．mostdevelopmental changes are the product of both maturation and learning．And as we willsee throughout this book．some of the more lively debates about human development arearguments about which of these processes contributes most to particular developmentalchanges．What GoaIs Do Developmentalists Pursue?

What objectives have developmentalists set for themselves?Three major goals stand outto describe，to explain，and to optimize development(Baltes，Reese，Lipsitt，1980)．Irpursuing the goal of description，human developmentalists carefully observe the behavior of people of different ages，seeking to specify how human beings change over timeThough there are typical pathways of development that virtually all people follow,researchers have discovered that no two persons are exactly alike．Even when raised in th(same home，children often display very different interests，values，abilities，and behaviors．Thus to adequately describe development，it is necessary to focus both on typicalpatterns of change(or normative development)and on individual variations(or ideographic development)，seeking to identify the important ways that developing humamresemble each other and how they are likely to differ as they proceed through life．

Adequate description provides us with the facts about development．but it is only thestarting point．Ultimately,developmentalists seek to explain the changes they have ob·served．In pursuing this goal of explanation，researchers hope to determine why humamdevelop as they typically do and why some individuals turn out differently from others．Stated another way,explanation centers both on normative changes within individuahand variations in development between individuals．As we will see throughout the text．jt is often easier to describe development than to conclusively explain why it occurs．

Finally,many researchers and practitioners hope to optimize development by applying what they have learned in attempts to help human beings develop in positive direc-tions．This is clearly a practical side to the study of human development that has led tostlch breakthrouluhs as ways t0·

Promote strong affectional ties between fussy,unresponsive infants and their frus-trated parents．

Assist children with learning difficulties to succeed at sch001．

Help socially unskilled children and adolescents to prevent the emotional difficultiesthat could result from having no close friends and being rejected by peers．

Many believe that such optimization goals increasingly will influence research agen-das in the twenty-first century(Fabes，Martin，Hanish，Updegraff，2000；Lerner,Fisher,Weinber9，2000)as developmentalists show greater interest in solving real problemsand communicating the practical implications of their findings to the public and to pol-icy fnakers(McCallGroark，2000)．Yet this heavier focus on applied issues in no wayimplies that traditional descriptive and explanatory goals are any less important becauseoptimization goals often cannot be achieved until researchers have adequately describednormal and abnormal pathways of development and their causes(Schwebel，Plumert，Pick，2000)．

Some Basic Observations about胁e Charactel。ofDevelopmentNow that we have defined development and talked briefly about the goals that develop．

mentalists pursue，let's consider some of the conclusions they have drawn about the char．acter of development．A Contillldt and Cumulative ProiTess

In his famous poemJ Paradise Lost,John Milton wrote：。。Childhood shows the man asmorning shows the day．This interesting analogy can be interpreted in at least two ways．It could be translated to mean that the events of childhood have little or no real impacton one's adult life，just as a sunny summer morning often fails to forecast an impendingafternoon thundershower．Yet most people do not interpret Milton's statement that way．Instead，they take it to mean that the events of childhood play a meaningful role in fore．casting the future．Human developmentalists clearly favor this latter interpretation．

Although no one can specify precisely what adulthood holds in store from even themost meticulous examination of a person's childhood，developmentalists have learnedthat the first l2 years are an extremely important part of the life span that sets the stagefor adolescence and adulthood．And yet how we perform on that stage also depends onthe experiences we have later in life．Obviously,you are not the same person you were atage lo or even at age l5．You have probably grown somewhat，acquired new academicskills，and developed very different interests and aspirations from those you had as afifth·grader or a high school sophomore．And the path of such developmental changestretches ever onward，through middle age and beyond，culminating in the final changethat occurs when we die．Altogether,human development is best described as a contin．ual and cumulative process．The one constant is change，and the changes that occur ateach major phase of life may have important implications for the future．

Table l．1 presents a chronological overview of the life span as developmentalists seeit．Our foCUS in this text is on development during the first five periods of life--theepochs known as childhood and adolescence．By examining how children develop fromthe moment they are conceived until they reach young adulthood．we willlearn moreabout ourselves and the determinants of our behavior．Oor survey will also provide someinsight as to why no two individuals are exactly alike，even when raised together in thesame home．1 won't promise that you will find answers to every important question youmay have about developing children and adolescents．The study of human developmentPeriod of rife Approximate age range

It was once fashionable to divide developmentalists into three camps：(1)those wh05tud。ied Dhysical growth and development，including bodily changes and the sequenmg otmotor skills；(2)those who studied cognitive aspects of development，including perceP。tion．1anguage．1earning，and thinking；and(3)those who concentrated on psycn050clalasDects of development，including emotions，personality,and the growth of mtooP0sonal relationships．Today we know that this classification is somewhat misleading，torresearchers who work in any of these areas have found that changes in one asPect of de。velop，ment have important implications for other aspects．Let's consider an example·

What determines a person's popularity with peers?If you were to say that soclal skillsare important you would be right．Social skills such as warmth，friendliness，and wl。In esj to cooperate are characteristics that popular children typically display．Yet theeis uch more to popularity than meets the eye．We now have some indication that theage at which a child reaches puberty,an important milestone in physical dev8lopm。nt，has an effect on social life．For example，boys who reach puberty early enjoy b8tt。1。l8tions with their peers than do boys who reach puberty later(LivsonPeskin，1980)Children who do well in school also tend to be more popular with their peers th8n chil-dren who perform somewhat less admirably in the classroom．

We see．then．that popularity depends not only on the growth of social skills but alsoon various aspects of both cognitive and physical development．As this example lllus。trates．development is not piecemeal but holistic--humans are physical，cognltl78，8ndsocial beings，and each of these components of self depends in part on chang。 t8kmgDlace in other areas of development．This holistic perspective is perhaps the domlnanttheme of human development today,around which this book is organlzedPlasticit)Plasticitv refers to a capacity for change in response to positive or negatlve llfe exPen。ences．Although we have described development as a continual and cumulatlve proc855and noted that past events often have implications for the future，developmentall5t5 naveknown for sOme time that the course of development can change abruptly if lmportantasDects of oHe，s life change．For example，somber babies living in barren，undestaffed orpanages often become quite cheerful and affectionate when placed in socially stimulat-ing ad云ptive homes(Rutter,1981)．Highly aggressive children who are intensely di．sli...ke．db，Deeoften improve their social status after learning and practicing the social s川5th。a；popular children display(MizeLadd，1990；Shure，1989)．It is indeed fortunatEthatan development is so plastic，for children who have horrible starts can often b￡helped to overcome their deficiencies Hish：jrical arm Cultural Context

No single portrait of development is accurate for all cultures，social classes，oraclaand ethnic groups．Each culture，subculture，and social class transmits a partlculaPattern of beliefs，values，customs，and skills to its younger generations，and the conten'of this cultural socialization has a strong influence on the attributes and competencle：that individuals display．Development is also influenced by societal changes：historical events such as wars，technological breakthroughs such as the development of P。sonal computers，and social causes such as the women's movement Each geneatl01develops in its own way,and each generation changes the world for succeeding gen．erations．So we should not automatically assume that developmental sequ8n。eS observed in samples of North American or European children(the most heavily studieDopulations)are optimal or even that they characterize persons developing1n otneeras or cultural settings(Laboratory for Comparative Human Cognition，1983)．Only：by adopting a historical／cultural perspective can we fully appreciate the richness anddiversity of human development．

Contemporary Western societies can be described aschild-centered：People often thinkof births as blessed events，spend a great deal of money to care for and educate theiryoung，and excuse children from shouldering the full responsibilities of adulthood untilattaining the legal age of l4 t021(depending on the society)，when they have presum．ably gained the wisdom and skills t0pull their own weight．Yet childhood and adoles．cence were not always regarded as the very special and sensitive periods that we regardthem as today．To understand how developmentalists think about and approach thestudy of children．it is necessary to see how the concept of childhood developed overtime．You may be surprised just how recent our modern viewpoint really is．Of course．itwas only after people came to view childhood as a very special period that they begin to；tudy children and the developmental process．

In the early days of recorded history,children had few if any rights，and their lives werenot always valued by their elders．Archeological research，for example，has shown thatthe ancient Carthaginians often killed children as religious sacrifices and embedded themin the walls of buildings t0strengthenthese structures(BjorldundBjorklund，1992)．Until the fourth century A．D．，Roman parents legally were entitled to kill their deformed，illegitimate，or otherwise unwanted infants．After this active infanticide was outlawed，unwanted babies were often left to die in the wilderness or were sold as servants or as ob．1ects for sexual exploitation on reaching middle childhood(deMause，1974)．Evenwanted children were often treated harshly by today's standards．For example，boys inthe city·state of Sparta were exposed to a strict regimen designed to train them for thegrim task of serving a military state．As infants．they were given cold baths t0toughen，，them．At age 7，when children in modern society are entering second grade，Spartan boyswere taken from their homes and housed in public barracks where they were often beatenor underfed to instill the discipline they would need to become able warriors(deMause．1974；Despert，1965)．

Not all early societies treated their children as harshly as the citizens of Carthage，Rome，and Sparta．Yet for several centuries after the birth of Christ，children were viewedas family possessions who had no rights(Hart，1991)and whom parents were free to ex．ploit as they saw fit．In fact，it wasn't until the twelfth century A．D．in Christian Europethat secular legislation equated infanticide with murder(deMause．1974)．

Currently,there is some debate about what childhood was like during the medievalera．Historian Philippe Aries(1962)has analyzed documents and paintings from medievalEurope and concluded that European societies had no concept of childhood as we know itbefore l600．Medieval children were not coddled or indulged to the extent that today，schildren are．They were often dressed in miniature versions of adult clothing and were de．picted in artwork working alongside adults in the shop or the field or drinking and carous．ing with adults at parties．And except for exempting very young children from criminal cul．pability,medieval law generally made no distinctions between childhood and adultoffenses(Borstelmann，1983；Kean，1937)．

Although medieva，children dressedfike their elders and often workedalongside them,it is doubtfut thatmwere considered miniature adf．But were medieval chil．dren really considered to beminiature adults?Probablynot．More recent and exten-sive examinations of me-dieval history reveal thatchildhood was generally rec-ognized as a distinct phase oflife and that children werethought to have certain needsabove and beyond those ofadults(Borstelmann，1983；Kroll，1977)．Clearly the ex-periences of children weredifferent during medievaltimes than today,for many

medieval children routinely performed economic activities that closely resemble adult：work by today'S standards．Yet，it is certainly an overstatement to conclude that medieval jsocieties had absolutely no concept of childhood and merely treated their young as：miniature adults(Cunningham，1996) bward Modern。Day Views 0摊Childhood During the seventeenth and eighteenth centuries，attitudes toward children and childrearing began to change．Religious leaders of that era stressed that children were innocentand helpless souls who should be shielded from the wild and reckless behavior of adults．One method of accomplishing this objective was to send young people to sch001．Al-though the primary purpose of schooling was to provide a proper moral and religious ed。ucation，it was now recognized that teaching important subsidiary skills such as readingand writing would transform the innocents into servants and workers who would pro-vide societywith a good labor force(Aries，1962，P．10)(see also Box l．1)．Althoughchildren were still considered family possessions，parents were nOW discouraged fromabusing their sons and daughters and were urged to treat them with more warmth andaffection(Aries，1962；Despert，1965)．

Why did attitudes toward children change SO drastically in the seventeenth and eigh-teenth centuries?Although the historical record is not very clear on this point，it is likelythat the thinking of influential social philosophers contributed meaningfully to the newlook at children and child care．Lively speculation about human nature led these philoso-phers to carefully consider each of the following issues：

1．Are children inherently good or bad?

2．Are children driven by inborn motives and instincts；oL rather,are they products oftheir environments?

3．Are children actively involved in shaping their characters；or are they passive crea'tures molded by parents，teachers，and other agents of society?

Debates about these philosophical questions produced quite different perspectiveson children and child rearing．For example，Thomas Hobbes's(1651／1904)doctrine oforiginal sin held that children are inherently selfish egoists who must be restrained bysociety,whereas Jean Jacques Rousseau's(1762／1955)doctrine of innate purity main-tained that children are born with an intuitive sense of right and wrong that society of-ten corrupts．These two viewpoints clearly differ in their implications for child rearing．Proponents of original sin argued that parents must actively control their egoistic off-spring while the innate purists viewed children asnoble savageswho should be givenmore freedom to follow their inherently positive inclinations．

Another influential view on children and child rearing was suggested by JohnLocke(1690／1913)，who believed that the mind of an infant is a tabula rasa，orblank slate，and that children have no inborn tendencies．In other words。childrenare neither inherently good nor inherently bad，and how they turn out depends en·tirely on their worldly experiences．Like Hobbes，Locke argued in favor of disciplinedchild rearing to ensure that children would develop good habits and acquire few ifany bad ones．

These philosophers also differed on the question of children's participation in theirown development．Hobbes maintained that children must learn to rechannel their natu-rally selfish interests into socially acceptable outlets；in this sense，they are passive sub-jects to be molded by the more powerful elements of society．-namely,parents．Locke，t00，believed that the child's role is passive because the mind of an infant is a blank slateon which experience writes its lessons．But a strikingly different view was proposed byRousseau，who believed that children are actively involved in the shaping of their ownintellects and personalities。In Rousseau's words，the child is not apassive recipient ofthe tutor's instructionbut abusy,testing。motivated explorer．The active searchingchild，setting his own problems，stands in marked contrast to the receptive one．．．onwhom society fixes its stamp(Kessen，1965，P．75)．

Clearly these philosophers had some interesting ideas about children and child rear-ing．But how could anyone decide whether their views were correct?Unfortunately,thephilosophers collected no objective data to back their pronouncements，and the few ob-servations they did make were limited and unsystematic．Can you anticipate the nextstep in the evolution of developmental sciences?

Childn'n aS Sufecls：The Baby Bio,graphies The first glimmering of a systematic study of children can be traced to the late l9th cen-tury．This was a period in which investigators from a variety of academic backgrounds be-gan to observe the development of their own children and to publish these data in worksknown as baby biographies．

Perhaps the most influential of the baby biographers was Charles Darwin，whomade daily records of the early development of his son(Darwin，1877；Charlesworth，1992)．Darwin's curiosity about child development stemmed from his earlier theory ofevolution．Quite simply,he believed that young，untrained infants share many charac-teristics with their nonhuman ancestors，and he advanced the(now discredited)ideathat the development of the individual child retraces the entire evolutionary history ofthe species。thereby illustrating thedescent of man．So Darwin and many of his con-temporaries viewed the baby biography as a means of answering questions about ourevolutionary past．

Baby biographies left much to be desired as works of science．Different baby biogra-phers emphasized very different aspects of their children's behavior,so different baby bi-ographies were difficult to compare．Furthermore，parents are not entirely objective abouttheir own children，and baby biographers like Charles Darwin may also have let their as-sumptions about the nature of development bias their observations so that they onlyfound what they were looking for．Finally,each baby biography was based on a singlechild--and often the child of a distinguished individual．Conclusions based on a singlecase may not hold true for other children．

Despite these shortcomings，baby biographies were a step in the right direction．Thefact that eminent scientists such as Charles Darwin were now writing about developingchildren implied that human development was a topic worthy of scientific scrutiny．

Introductory textbooks in most academic areas typically credit someone as the founderof the discipline．In developmental psychology,there were several influential pioneerswho merit consideration for this honor．Still，the person who is most often cited as thefounder of developmental psychology is G．Stanley Hall．

Well aware of the shortcomings of baby biographies，Hall set out in the late l9thcentury to collect more objective data on larger samples．Specifically,he was interestedin children's thinking。and he developed a now familiar research tool-the question-naire-to explorethe contents of children's minds(Hall，1891)．By asking childrenquestions about a range of topics，Hall discovered that children's understanding of theworld grows rapidly during childhood and that the logic of young children is not verylogical at all．Hall later wrote an influential book titled Adolescence(1904)that was thefirst work to call attention to adolescence as a unique phase of the human life span．Hiswork was the first large-scale scientific investigation of developing youth，and it is onthis basis that G．Stanley Hall merits consideration as the founder of developmental psy-chol0胛(White，1992)．

At about the time Hall was using questionnaires to study children's minds，a youngEuropean neurologist was trying a different method of probing the mind and revealingits contents．This neurologist's approach was very fruitful，providing information that1ed him to propose a theory that revolutionized thinking about children and childhood．The neurologist was Sigmund Freud，and his ideas came to be known as psychoanalytictheory．

In many areas of science，new theories are often revisions or modifications of old the-ories．But in Freud's day,there were few old theories of human development to modify．Freud was truly a pioneer,formulating his psychoanalytic theory from the thousands ofnotes and observations he made while treating patients for various kinds of emotionaldisturbances．

Freud's highly creative and unorthodox theorizing soon attracted a lot of attention．Shortly after the publication of Freud's earliest theoretical monographs．the InternationalIournal of Psychoanalysis was founded，and other researchers began to report their testsof Freud's thinking．By the mid l930s．much of Freud's work had been translated intoother languages，and the impact of psychoanalytic theory was felt around the world．Over the years，Freud's theory proved to be quite heuristic，meaning it continued to gen。erate new research and to prompt other researchers to revise and extend Freud's think．ing．Clearly,the field of developmental psychology．was thriving by the time Freud diedin l939．

Freud's work-and other scientists'reactions to it-aDtlv illustrates the role that the．ories play in the science of human development．Although the word theory is an impos-ing term．theories are Something that everybody has．If l were to ask you why males andfemales appear different as adults when they seem so very similar as infants，you proba-blv would have some opinions on the issue．Your answer would state or at least refectyour own underlying theory of sex differences．So a theory is nothing more than a set ofconcepts and propositions that allow the theorist to describe and explain some aspect ofexperience．In the field of psychology,theories help us to describe and explain variouspatterns of behavior．

Good theories have another important feature：the ability to predict future events．These theoretical predictions，or hypotheses，are then tested by collecting additionaldata．The information we obtain when testing hypotheses not only provides some cluesabout the theory's ability to explain new observations but may also lead to new theoret-ical insights that extend our knowledge．

Today there are many theories that have contributed to our understanding ofchild and adolescent development，and in Chapter 2．we will examine several of themore influential of these viewpoints．Although it is natural for people reading aboutvarious theories to favor one，the scientist uses a rather stringent yardstick to evaluatetheories：He or she will formulate hypotheses and conduct research to determinewhether the theory can adequately predict and explain new observations．Thus thereis no room for subjective bias when evaluating a theory．Theories in the developmen-tal sciences are only as good as their ability to predict and explain important aspectsof development．

In the next section of the chapter,we will foCUS on the tools of the trade-that is．theresearch methods that developmentalists use to test their theories and gain a better un-derstanding of child and adolescent development．

When detectives are assigned cases to solve，they first gather the facts，formulate hunche!and then sift through the clues or collect additional information until one of thejhunches proves correct．Unraveling the mysteries of development is in many ways a sireilar endeavor．Investigators must carefully observe their subjects，analyze the informatio Jthey collect，and use these data to draw conclusions about the ways people develop．

0ur foCUS in this section is on the methods that researchers use to gather information about developing children and adolescents．0ur first task is to understand why developmentalists consider it essential to collect all these facts．We will then discuss the advantages and disadvantages of five basic fact-finding strategies：self-report methodologie!systematic observation，case studies，ethnography,and psychophysiological methods．Finally,we will consider the ways developmentalists might design their research to detecand explain age，related changes in children's feelings，thoughts，abilities，and behavior!

Modern developmental psychology is appropriately labeled a scientific enterprise becausethose who study developing organisms have adopted a value system we call the scien-tific method that guides their attempts at understanding．There is nothing mysteriousabout the scientific method．It is really more of an attitude or value than a method；onewhich dictates that，above all，investigators must be objective and must allow their ob-servations for data)to decide the merits of their thinking．

In earlier eras．when social philosophers such as Hobbes，Locke，and Rousseau werepresenting their views on children and child rearing，their largely unsubstantiated claimswere often accepted as fact．People assumed that great minds always had great insights．

Few individuals questioned the word of these well-knownscholars because the scientific method was not yet awidely accepted criterion for evaluating knowledge．

The intent here is not to criticize the early socialphilosophers．In fact．today's developmentalists(and chil·dren)are deeply indebted to these thinkers for helping tomodify the ways in which society regarded and treated itsyoung．However,great minds may on occasion produce

miserable ideas that can do a great deal of harm if their errors in thinking are uncriticallyaccepted and influence the way human beings are treated．The scientific method，then，isa valuable safeguard that helps to protect the scientific community and society at largeagainst flawed reasoning．Protection is provided by the practice of evaluating the meritsof various theoretical pronouncements against the objective record rather than simply re。lvlng on the academic，political，or social credibility of the theorist．Of course，this alsomeans that the theorist whose ideas are being evaluated must be equally objective andthus willing to discard pet notions When there is evidence against them．

No matter what aspect of development we hope to studye it the perceptual capabili·ties of newborn infants，the growth of friendships among grade school children，or thereasons some adolescents begin to use drugs--we must find ways to measure what inter。

ests us．Today researchers are fortunate in having many tried-and．true procedures thatthey might use to measure behavior and to test their hypotheses about human develop．ment．But regardless of the technique one employs，scientifically useful measures mustalways display two important qualifies：rellabilitv and validitv．

A measure is reliable if it vields consistent information over time and across ob．servers．Suppose you go into a classroom and record the number of times each child be．haves aggressively toward others，but your research assistant，using the same scheme toobserve the same children，does not agree with your measurements；or you measure eachchild's aggressiveness one week but come up with very different aggressiveness scoreswhile applying the same measure to the same children a week later．Clearly,your obser·vational measure of aggression is unreliable because it yields highly inconsistent infor．mation．T0 be reliable and thus useful for scientific purposes．your measure would haveto produce comparable estimates of children's aggression from independent observers(interrater reliability)and would yield similar scores for individual children from onetesting to another shortly thereafter ftemporal stability)．

A measure is valid if it measures what it is supposed to measure．Obviously,an in．strument must be reliable and measure consistently before it can possibly be valid．Yet re．1iabili劬by itself,does not guarantee validity(Miller,1997)．For example，a highly reli-able observational scheme that is intended as a measure of children's aggression mayprovide grossly overinflated estimates of aggressive behavior if the investigator simplyclassifies all acts of physical force as examples of aggression．What the researcher hasfailed to recognize is that much high-intensity behavior simply may represent enjoyableforms of rough-and-tumble play without harmful or aggressive intent．Clearly，re-searchers must demonstrate that they are measuring the attribute they say they are mea．suring before we can have much faith in the data they collect or the conclusions theyreach．

Keeping in mind the importance of establishing the reliability and validity of mea-sures．1et us consider some of the different ways in which aspects of human developmentmight be measured．

Yhree common procedures that developmentalists use to gather information and test hy．potheses are interviews，questionnaires(including psychological tests)，and the clinicalmethod．Although these approaches are similar in that each asks participants to answerquestions posed by the investigator,they differ in the extent to which the investigatortreats individual participants alike．

Researchers who opt for interview or questionnairetechniques will ask the child(or the child's parents)a series of questions pertaining tosuch aspects of development as the child's conduct，feelings，beliefs，or characteristicmethods of thinking．Collecting data via a questionnaire(and most psychological tests)simply involves putting questions on paper and asking participants to respond to themin writing，whereas interviews require participants to respond orally to the investigator'squeries．If the procedure is a structured interview or structured questionnaire．all whoparticipate in the study are asked the same questions in the same order．The purpose ofthis standardized or structured format is to treat each person alike so that the responsesof different participants can be compared．

One interesting use of the interview technique is a project in which kindergarten，second-grade，and fourth-grade children responded t024 questions designed to assesstheir knowledge of social stereotypes about males and females(Williams，Bennett．Best．1975)．Each question came in response to a different short story in which the centralcharacter was described by either stereotypically masculine adjectives(for example，ag-gressive，forceful，tough)or stereotypically feminine adjectives(e．g．，emotional，ex．citable)．The child's task was to indicate whether the character in each story was male orfemale．Williams and associates found that even kindergartners could usually tellwhether the stories referred to boys or girls．In other words。these 5-year-olds were quiteknowledgeable about gender stereotypes although children's thinking became more stereotyped between kindergarten and the second grade．One implication of these results is that stereotyping of thesexes must begin very early if kindergartners are alreadythinking along stereotyped lines．

Interviews and questionnaires have some real shortcom．ings．First，neither approach can be used with very youngchildren who cannot read or comprehend speech very well．Investigators also must hope that the answers they receiveare honest and accurate and are not merely attempts by re．spondents to present themselves in a favorable or sociallydesirable way．Many adolescents，for example，may be un．willing to admit that they regularly masturbate，that theysmoke marijuana，or that they enjoy the risks of shoplifting．Clearl％inaccurate or untruthful responses lead to erroneousconclusions．InvestigatOrs must also be careful to ensure thatparticipants of all ages interpret questions in the same way；otherwise the age trends observed in the study may reflectdifferences in children's ability to comprehend and commu．nicate rather than real underlying changes in their feelings，thoughts．or behaviors．Finally,researchers who interviewboth developing children and their parents(or teachers)

may have trouble determining which set of reports is more accurate should the children'sdescriptions of their own behaviors differ from those of the other informants．

Despite these potential shortcomings，structured interviews and questiOnnaires canbe excellent methods of obtaining large amounts of useful information in a short periodof time．Both approaches are particularly useful when the investigator emphasizes to par-ticipants that their responses will be confidential and challenges them to report exactlywhat they know about an issue，thereby maximizing the likelihood of a truthful or accu-rate answer．In the gender stereotyping study,for example，the young participants prob-ably considered each question a personal challenge or a puzzle to be solved and were thusmotivated to answer accurately and to display exactly what they knew about males andfemales．Under the circumstances，then，the structured interview was an excellentmethod of assessing children's perceptions of the sexes．

The clinical method is a close relative of the interview technique．The investigator usually is interested in testing a hypothesis by presenting the researchparticipant with a task or stimulus of some sort and then inviting a response．After the par-ticipant responds，the investigator typically asks a second question or introduces a newtask to clarify the participant's original answer．Although subjects are often asked the samequestions initially,each participant's answer determines what he or she is asked next．Thus，the clinical method is a flexible approach that considers each subject to be unique．Jean Piaget，a famous Swiss psychologist，relied extensively on the clinical method tostudy children's moral reasoning and intellectual development．The data from Piaget's re-search are largely protocol records of his interactions with individual children．Here is asmall sample from Piaget's work(1932／1965，P．140)on the development of moral rea-soning，which shows that this young child thinks about lying in a different way thanadults amounts of information in relatively brief periods．Proponents of this approach also citeits flexibility as an advantage：By asking follow-up questions that are tailored to the par-ticipant's original answers(as Piaget did in the above example)，it is often possible to ob-tain a rich understanding of the meaning of those answers．However,the flexibility of theclinical method is also a potential shortcoming．Consider that it may be difficult．if notimpossible，to directly compare the answers of participants who are asked different ques．tions．Furthermore，the practice of tailoring one's questions to the participant's responsesraises the possibility that the examiner's preexisting theoretical biases may affect the par．ticular follow．up questions asked and the interDretations provided．Because conclusionsdrawn from the clinical method depend in part on the investigator's subjective interpre．tations．it is always desirable to verify these insights using other research techniques．

often researchers prefer to observe people's behavior directly rather than asking them questions about it．One method that many developmentalists favor is naturalistic obser．vation---observing people in their common(i．e．，natural)surroundings(Pellegrini，1996)．To observe children，this usually means going into homes，schools，or public parks andplaygrounds and carefully recording what they d0．Rarely will the investigator try torecord every event that OCCURS；they usually are testing a specific hypothesis about onetype of behavior,such as cooperation or aggression，and will focus exclusively on acts ofthis kind．One strength of naturalistic observation is the ease with which it can be appliedto infants and toddlers．who often cannot be studied through methods that demand ver．bal skills．But perhaps the greatest advantage of the naturalistic observation is that .

However,naturalistic observation also has its limitations．First．some behaviors occurso infrequentlyor are so socially undesirable play that they are unlikely to be witnessed by an unknown observer in the naturalenvironment．Second．many events are usually happening at the same time in a naturalsetting，and any(or some combination)of them may affect people's behavior．This makesit difficult to pinpoint the causes of participants'actions or of any developmental trendsin behavior．Finally,the mere presence of an observer can sometimes make people behavedifferently than they otherwise would．Children may ham it up when they have an an．dience，whereas parents may be on their best behavior，showing a strong reluctance，forexample，to spank a misbehaving child as they normally might．For these reasons，re-searchers often attempt to minimize observer influence by fl)videotaping their partici．pants from a concealed location or(2)spending time in the setting before collecting theirreal data so that the individuals they are observing will grow accustomed to their pres．ence and behave more naturally．

Mary Haskett and lanet Kistner(1991)conducted an excellentpiece of naturalistic observation to compare the social behaviors of nonabused presch001．ers with those of day-care classmates identified by child protection agencies as havingbeen physically abused by their parents．The investigators first defined examples of thebehaviors they wished to record--both desirable behaviors．such as appropriate socialini．tiations and positive play,and undesirable behaviors，such as aggression and negativeverbalizations．They then monitored l4 abused and l4 nonabused preschool children asthey mingled with peers in a play area of a day．care facility．Observations were made ac．cording to a time-sampling procedure：each child was observed during three l0-minuteplay sessions on three different days．To minimize their influence on the play activities，observers stood outside the play area while making their observations．

The results were disturbing As shown in Figure l．1．abused children initiated fewersocial interactions than their nonabused classmates and were somewhat socially with．drawn．When they did interact with playmates，the abused youngsters displayed more a9-gressive acts and other negative behaviors than did their nonabused companions．Indeed．nonabused children often blatantly ignored the positive social initiations of an abusedchild as if they did not want to get involved with him or her．

In sum．Haskett and Kistner's observational study shows that abused children are on．attractive playmates who are likely to be disliked and even rejected by peers．But as is al．most always the case in naturalistic observational research．it is difficult to pinpoint theexact cause of these findings．Did the negative behaviors of abused children cause theirpeers to reject them?Or did the peer rejection cause the abused children to display ne9．ative behaviors?Either possibility can account for Haskett and Kistner's results．

How might observationalresearchers study unusual orundesirable behaviors thatthey are unlikely to observe inthe natural environment?Oneway is to conduct structuredobservations in the labora．torⅥIn a structured observa．tional study,each participantis exposed to a setting thatmight cue the behavior in

question and is then surreptitiouslv observed(via a hidden camera or through a one-waymirror)to see if he or she performs the behavior．For example，Leon Kuczynski(1983)got children to promise to help him with a boring task and then left them alone to workin a room where attractive toys were present．This procedure enabled Kuczynski to de-termine whether youngsters would break a promise to work(an undesirable act thatmany children displayed)when they thought there was no one present to observe theirtransgression．

Aside from being a most feasible way of studying behaviors that occur infrequentlyor are not openly displayed in the natural environment，structured observations also en-sure that every participant in the sample is exposed to the same eliciting stimuli and hasan equal opportunity to perform the target behavior--circumstances that are not alwaystrue in the natural environment．Of course．the major disadvantage of structured obser-vation is that participants may not always respond in a contrived laboratory setting asthey would in everyday life．

Any or all of the methods we have discussed--structured interviews，questionnaires，clinical methods．and behavioral observations--can be used to compile a detailedportrait of a single individual's development through the case study method．Inpreparing an individualized record，or case，the investigator typically seeks many kindsof information about the participant，such as his or her family background，socioeco-nomic status，health records，academic or work history,and performance on psycho-logical tests．Much of the information included in any case history comes from inter-views with and observations of the individual although the questions asked andobservations made are typically not standardized and may vary considerably from caseto case．

The baby biographies of the nineteenth and early twentieth centuries are examplesof case studies，and Sigmund Freud conducted many fascinating case studies of his clini-cal patients．In analyzing his cases，Freud noticed that different patients often describedvery similar events and experiences that had been noteworthy to them as they weregrowing up．He inferred from the observations that there must be important milestonesin human development that all people share．As he continued to observe his patients andt01isten to accounts of their lives，Freud concluded that each milestone in the life his-tory of a patient was meaningfully related to earlier events．He then inferred that he hadthe data to construct a comprehensive explanation of human development--the accountwe know today as psychoanalytic theory．

Although Freud and many other developmentalists have used case studies to greatadvantage，there are major drawbacks to this approach．For example，it is often difficultto directly compare cases who have been asked different questions，taken different tests，and been observed under different circumstances．Case studies may also lack generaliz-ability；that is，conclusions drawn from the experiences of the small number of individ-uals studied may simply not apply to most people．In fact，one recurring criticism ofFreud，s psychoanalytic theory is that it was fcIrmulated from the life histories of emo-tionally disturbed patients who are hardly typical of the general population．For thesereasons，any conclusions drawn from case studies should always be verified through theuse of other research techniques．

Ethnography--a form of participant observation often used in thefield of anthropology--is becoming increasingly popular among re-searchers who hope to understand the impacts of culture on devel．oping children and adolescents．To collect their data，ethnogra-phers often live within the cultural or subcultural community theyare studying for periods of months，or even years．The data they col·lect is typically diverse and extensive，consisting largely of natural-istic observations．notes made from conversations with members ofthe culture，and initial interpretations of these events．These dataare eventually used to compile a detailed portrait of the culturalcommunity and to draw conclusions about how the community'sunique values and traditions influence one or more aspects of thedevelopment of its children and adolescents．

Clearly detailed ethnographic portraits of a culture or subcul-ture that arise from close and enduring contact with members of

the community can lead to a richer understanding of that community's traditions andvalues than is possible through a small number of visits，in which outsiders make limitedobservations and conduct a few interviews(LeVine et al．，l994)．In fact，such extensivecultural or subcultural descriptions are particularly useful to investigators hoping tounderstand cultural conflicts and other developmental challenges faced by minoritychildren and adolescents in diverse multicultural societies(Segal，1991；Patel，Power,Bhavnagri，1996)．Despite these clear strengths，ethnography is a highly subjectivemethod because researchers'own cultural values and theoretical biases can cause them tomisinterpret what they have experienced．In addition，ethnographic conclusions pertainonly to the culture or subculture studied and cannot be assumed to generalize to othercontexts or social groups．

In recent years，developmentalists have turned to psychophysiological methods--tech-niques that measure the relationship between physiological responses and behavior--toexplore the biological underpinnings of children's perceptual，cognitive，and emotionalresponses．Psychophysi0109ical methods are particularly useful for interpreting the men-tal and emotional experiences of infants and toddlers who are unable to report suchevents(Bornstein，1992)

Heart rate is an involuntary physiological response that is highly sensitive to one'spsychological experiences．Compared to their normal resting，or baseline，levels，infantswho are carefully attending to an interesting stimulus may show a decrease in heart rate，whereas those who are uninterested in it may show no heart rate change，and others，who are wary of or angered by the stimulus，may show a heart rate increase(Campos，Bertenthal，Kermoian，1992；FoxFitzgerald，1990)．Measures of brain function arealso useful for assessing psychological state．For example，electroencephalogram(EEG)recordings of brain wave activity can be obtained by attaching electrodes to the scalp．Be-cause different patterns of EEG activity characterize different arousal states，such as sleep，drowsiness，and alertness，investigators can track these patterns and determine how sleepcycles and other states of arousal change with age．Novel stimuli or events also produceshort．term changes in EEG activity．So an investigator who hopes to test the limits of in-fant sensory capabilities can present novel sights and sounds and look for changes inbrain waves(called event·related potentials，or ERPs)to determine whether these stimulihave been detected，or even discriminated，for two stimuli sensed as different will pro-duce different patterns of brain activity(Bornstein，1992)．

Though very useful，psychophysiological responses are far from perfect indicators ofpsychological states．Even though an infant's heart rate or brain wave activity may indi-cate that he or she is attending to a stimulus，it is often difficult to determine exactlywhich aspect of that stimulus(shape，color,etc．)has captured attention．Furthermore，changes in physiological responses often reflect mood swings，fatigue，hunger,or even Method Strengths Limitattons.

Interviews and questionnaires Relatively quick way to gather much informa．Data collected may be inaccurate or less thantion；standardized format allows the investiga．completely honest，or may reflect variationstor to make direct comparisons between data in respondents'verbal skills and ability to

provided by different participants． understand questions．

Clinical methods Flexible methodology that treats subjects as Conclusions drawn may be unreliable in thatunique individuals；freedom to probe can be participants are not all treated alike；flexiblean aid in ensuring that the participant under-probes depend，in part，on the investigator'sstands the meaning of the questions asked．subjective interpretations of the participant'sresponses；can be used only with highly ver．bal participants．

Systematic observations Naturalistic observation Allows study of behavior as it actually occurs Observed behaviors may be influenced by in the natural environment．observer's presence；unusuaI or undesirable behaviors are unlikely to be observed duringthe periods when observations are made．

Structured observation Offers a tandardized environment that pr0． Contrived observations may not always cap．vides every child an opportunity to perform ture the ways children behave in the naturaltarget behavior．Excellent way to observe environment．infrequent or socially undesirable acts．Case Studies Very broad method that considers many Kind of data collected often differs from casesources of data when drawing inferences and to case and may be inaccurate or less than

conclusions about individual participants．honest；conclusions drawn from jndividualcases are subjective and may not apply toother people．

Ethnography Provides a richer description of cultural beliefs， Conclusions may be biased by the investiga．

values，and traditions than is possible in brief tor's values and theoretical viewpoints；resultsobservational or interview studies． cannot be generalized beyond the groups andsettings that were studied．

Psychophysiological Useful for assessing biological underpinnings Cannot indicate with certainty what partici-methods of development and identifying the percep-pants sense or feel；many factors other thantions，thoughts，and emotions of infants and the one being studied can produce a similartoddlers who cannot report them verbally．physiological response．

negative reactions to the physiological recording equipment rather than a change in theinfant's attention to a stimulus or emotional reactions to it．For these reasons．physi0．109ical responses are more likely to be valid indications of psychological experience5when participants(particularly very young ones)are initially calm，alert，and contented．Table l．2 provides a brief review of the data-gathering methods that we have exam．ined thus far．In the sections that follow,we will consider how investigators might designtheir research to test hypotheses and detect developmental continuities and changes．

Once researchers have decided what they want to study,they must then devise a researchplan，or design，that permits them to identify relationships among events and behaviorsand to specify the causes of these relationships．

Here we consider the two general research designs that investigators might employ：correlational and experimental．

In a correlational design，the investigator gathers information to determine whethertwo or more variables of interest are meaningfully related．If the researcher is testing aspecific hypothesis(rather than conducting preliminary descriptive or exploratory re-search)．he or she will be checking to see whether these variables are related as the hy．pothesis specifies they should be．No attempts are made to structure or to manipulate theparticipants'environment in any wav Instead。correlational researchers take people asthey find them--alreadymanipulatedby natural life experiences--and try to deter．mine whether variations in people's life experiences are associated with differences intheir behaviors or patterns of development．

To illustrate the correlational approach to hypothesis testing，let's work with a sim·ple theory specifying that youngsters learn a lot from watching television and are apt toimitate the actions of the characters they observe．One hypothesis we might derive fromthis theory is that the more frequently children observe TV characters who display vi0．1ent and aggressive acts，the more inclined they will be to behave aggressively towardtheir own playmates．After selecting a sample of children to study,our next step in test-ing our hypothesis is to measure the two variables that we think are related．To assesschildren's exposure to violent themes on television．we might use the interview or natu．ralistic observational methods to determine what each child watches．and then count thenumber of aggressive acts that occur in this programming．To measure the frequency ofthe children's own aggressive behavior toward peers，we could observe our sample on aplayground and record how often each child behaves in a hostile，aggressive manner t0-ward playmates．Having now gathered the data，it is time to evaluate our hypothesis．

The presence(or absence)of a relationship between variables can be determined bysubjecting the data to a statistical procedure that yields a correlation coefficient(sym．bolized by an rl．This statistic provides a numerical estimate of the strength and the di．rection of the relationship between two variables．It can range in value from+1．00 to1．o0．The absolute value of r(disregarding its sign)tells us the strength of the relation-ship．Thus correlation coefficients of一．70 and+．70 are of equal strength．and both arestronger than a moderate correlation of．50．An r of．00 indicates that the two variablesare not systematically related．The sign of the correlation coefficient indicates the direc．tion of the relationship．If the sign if positive，this means that as the variable increases．the other variable also increases．For example，height and weight are positively corre-lated：as children grow taller，they tend to get heavier(Tanner,1990)．Negative correla-tions，however．indicate inverse relationships；as one variable increases，the other de．creases．Among grade school students，for example，aggression and popularity arenegatively correlated：Children who behave more aggressively tend to be less popularwith their peers(Crick．1996)．

Now let's return to our hvDOthesized positive relationship between televised violenceand children's aggressive behavior．A number of investigators have conducted correla．tional studies similar to the one we have designed，and the results(reviewed in Liebertamount ofviolence that children see on television and the number ofaggressive responses display．EⅡch dot represents a specific childwho views a particular level oftelevised violence(shown on the horizontal axis)and commits a particular number ofaggressive acts(shown on the vertical axis)．Although the correlation less than perfe we see that the more acts ofviolence a child watches on TV,the more inclined he or she is to behave aggressively towardpeers．Sprafkin，1988)suggest a moderate positive correlation

rbetween+．30 and+．50)between the two variables ofinterest：Children who watch a lot of violent televisionprogramming are more likely to behave aggressively to-ward playmates than are other youngsters who watchlittle violent programming fsee Figure l．2 for a visualdisplay)．

Do these correlational studies establish that expo．sure to violent TV programming causes children to be．

have more aggressively?N0．they do not!Although we have detected a relationship be-tween exposure to televised violence and children's aggressive behavior,the causaldirection of the relationship is not at all clear．An equally plausible alternative explana．tion is that relatively aggressive children are more inclined to prefer violent program-ming．Another possibility is that the association between TV viewing and aggressive be-havior is actually caused by a third variable we have not measured．For example．perhapsparents who fight a lot at home fan unmeasured variable)cause their children to becomemore aggressive and to favor violent TV programming．If this were true．the latter twovariables may be correlated even though their relationship to each other is not one ofcause and eftect．

In conclusion．the correlational design is a versatile approach that can detect sys．tematic relationshiDs between any two or more variables that we might be interested inand are capable of measuring．However,its major limitation is that it cannot unambigu-ously indicate that one thing causes another．How,then．might a researcher establish theunderlying causes of various behaviors or other aspects of human development?0ne s0．1ution is to conduct experiments．

In contrast to correlational studies，experimental designs permit a precise assessment ofthe cause-and·effect relationship that may exist between two variables．Let's return to theissue of whether viewing violent television programming causes children to becomemore aggressively inclined．In conducting a laboratory experiment to test this(or any)hypothesis，we would bring participants to the lab，expose them to different treatments，and record their responses to these treatments as data．

The different treatments to which we expose our participants represent the inde-pendent variable of our experiment．To test the hypothesis we have proposed，our in-dependent variable(or treatments)would be the type of television program that our par-ticipants observe．Half the children might view a program in which characters behave ina violent or otherwise aggressive manner toward others，whereas the other half wouldwatch a program that contains little if any violence．

Children's reactions to the television shows would become the data，or dependentvariable，in our experiment．Because our hypothesis centers on children's aggression，wewould want to measure(as our dependent variable)how aggressively children behave af-ter watching each type of television show．A variable is dependent because its value pre-sumably depends on the independent variable．In the present case，we are hypothesizingthat future aggression(our dependent variable)will be greater for children who watch vi-olent programs fone level of the independent variablel than for those who watch nonvi．

Olent programs(the second level of the independent variable)．If we are careful experi。menters and exercise precise control over all other factors that may affect children's ag-gression．then the pattern of results that we have anticipated will allow us to draw astrong conclusion：Watching violent television programs causes children to behave moreaggressively．

An experiment similar to the one we have proposed was actually conducted(LiebertBaron．1972)．Half the 5．t09-．year．olds in this study watched a violent 3一minute clipfrom the Untouchables--one that contained two fistfights，two shootings，and a stabbing．The remaining children watched a 3．minute film of a nonviolent but exciting track meet．So the independent variable was the type of program watched．Then each child was takeninto another room and seated before a panel that had wires leading into an adjoiningroom．On the panel was a green button labeled HELE a red button labeled HURT,and awhite light between the buttons．The experimenter then told the child that another childin the adjoining room would soon be playing a handle-turning game that would illumi。nate the white light．The participant was told that by pushing the buttons when the lightwas lit，he or she could either help the other child by making the handle easy to turn orhurt the child by making the handle become very hot．When it was clear that the par-ticipant understood the instructions，the experimenter left the room，and the light cameon 20 times over the next several minutes．So each participant had 200pportunities tohelp or hurt another child．The total amount of time each participant spent pushing thehurt button served as a measure of his or her aggression--the dependent variable in thisstudy．

The results were clear：Despite the availability of an alternative，helping response，both boys and girls were much more likely to press the HURT button if they had watchedthe violent television program．So it appears that a mere 3-minute exposure to televisedviolence can cause children to behave more aggressively toward a peer，even though theaggJressive acts they witnessed on television bore no resemblance to those they commit-ted themselves．

When students discuss this experiment in class，someone invariably challenges thisinterpretation of the results．For example，one student recently proposed an alternativeexplanation thatmaybe the kids who watched the violent film were naturally more a9-gressive than those who saw the track meet．In other words，he was suggesting that aconfounding variable--children's preexisting levels of aggression--had determinedtheir willingness to hurt a peer and that the independent variable(type of television pro-gram)had had no effect at all．Could he have been correct?How do we know that thechildren in the two experimental conditions didn't differ in some important way thatmay have affected their willingness to hurt a peer?

This question brings us to the crucialissue of experimental contr01．In order to con-clude that the independent variable is causally related to the dependent variable．the ex-perimenter must ensure that all other confounding variables that could affect the de-pendent variable are controlled-that is．equivalent in each experimental condition．Oneway to equalize these extraneous factors is to do what Liebert and Baron(1972)did：Ran-domly assign children to their experimental treatments．The concept of randomization，or random assignment，means that each research participant has an equal probability ofbeing exposed to each experimental treatment or condition．Assignment of individualparticipants to a particular treatment is accomplished by an unbiased procedure such asthe flip of a coin．If the assignment is truly random，there is only a very slim chance thatparticipants in the two(or more)experimental conditions will differ on any characteris-tic that might affect their performance on the dependent variable；all these confoundingvariables will have been randomly distributed within each condition and equalized acrossthe different conditions．Because Liebert and Baron randomly assigned subjects to exper-imental conditions．they could be reasonably certain that children who watched the vi-olent TV program were not naturally more aggressive than those who watched the non-violent TV program．So it was reasonable for them to conclude that the former group ofchildren were more aggressive because they had watched a TV program in which violenceand aggression were central．

The greatest strength of the experimental method is its ability to establish unam biguously that one thing causes another．Yet，critics of laboratory experimentation have we consider a step that experimentalists can take to counter this criticism and assess theecological validity of their laboratory findings．

There are many issues to which an experimental design either cannot be applied orshould not be used for ethical reasons．Suppose，for example，that we wish to study theeffects of social deprivation in infancy on children's intellectual development．Clearly wecannot ask one group of parents to lock their infants in an attic for 2 years so that we cancollect the data we need．It is simply unethical to subject children to any experimentaltreatment that would adversely affect their physical or psychological well．being．

However,we might be able to accomplish our research objectives through a natural(or quasi-)experiment in which we observe the consequences of a natural event that sub．iects have experienced．So if we were able t010cate a group of children who had been raisedin impoverished institutions with limited contact with caregivers over the first two years．we could compare their intellectual development with that of children raised at home withtheir families．This comparison would provide valuable information about the likely im．pact of early social deprivation on children's intellectual development．rIndeed．preciselythis kind of natural experiment is described in detailin Chapter l l．The independent vari．able in a natural experiment is the event that subjects experience fin our example．the s0．cial deprivation experienced by institutionalized infants)．The dependent variable is what．ever outcome measure one chooses to study(in our example．intellectual development)．Let's note，however,that researchers conducting natural experiments do not controlthe independent variable，nor do they randomly assign participants to experimental con．ditions,instead，they merely observe and record the apparent outcomes of a natural hap．pening or event．And in the absence of tight experimental contr01．it is often hard to de．termine precisely what factor is responsible for any group differences that are found．Suppose，for example，that our socially deprived institution children showed a poorerpattern of intellectual outcomes than children raised at home．Is the social deprivationthat institutionalized children experienced the factor that accounts for this difference?Or is it that institutionalized children differed in other ways from family．reared children(e．g．，were more sickly as infants，were more poorly nourished．or simply had less intel．1ectual potential)that might explain their poorer outcomes?Without randomly assign．ing participants to treatments and controlling other factors that may vary across treat．Ments we simply cannot be certain that social deprivation is thefactor responsible for the poor intellectual outcomes that institutionalized children dis．play,

Despite its inability to make precise statements about cause and effect，the naturalexDeriment is useful because it can tell us whether a natural event could possibly haveinfluenced those who experienced it and thus can provide some meaningful clues aboutcause and effect．

To test these hypotheses．Coates and Hartup de．signed an interesting cross．sectional experiment．Chil．dren from two age groups一4一t05-year-olds and 7．to8．year．olds--watched a short film in which an adultmodel displayed 20 novel responses．such as throwinga beanbag between his legs，lassoing an inflatable toywith a Hula Hoop．and so on．Some of the childrenfrom each age group were instructed to describe themodel's actions，and they did so as they watched thefilm(induced．verbalization condition)．Other childrenwere not required to describe the model's actions asthey observed them fpassive．observation condition)．When the show ended，each child was taken to a roomthat contained the same toys seen in the film and wasasked to demonstrate what the model had done withthese toys．

An important advantage of the cross-sectional design is that the investigator can c01．1ect data from children of different ages over a short time．For example，Coates andHartup did not have to wait 3 years for their 4．t05-year．olds to become 7．t08．year．oldsin order to test their developmental hypotheses．They merely sampled from two agegroups and tested both samples simultaneously．Yet there are two important limitationsof cross．sectional research．

Notice that in cross-sectional research．participants at each age level are different people．That is，they come from different cohorts，where a cohort is defined as a group of peopleof the same age who are exposed to similar cultural environments and historical eventsas they are growing up．The fact that cross-sectional comparisons always involve differ．ent cohorts presents us with a thorny interpretive problem，for any age differences thatare found in the study may not always be due to age or development but，rather,may re．flect other cultural or historical factors that distinguish members of different cohorts．Stated another way,cross．sectional comparisons confound age and cohort effects．

An example should clarify the issue．For years，cross．sectional research had consis．tently indicated that young adults score slightly higher on intelligence tests than middle-aged adults，wh0，in turn，score much higher than the elderly．But does intelligence cline with age，as these findings would seem to indicate?Not necessarily．Morerecentresearch(Schaie1990)reveals that individuals'intelligence test scores remain relativelystable over the years and that the earlier studies were really measunng something quitedifferent：age differences in education．The older adults in the cross-sectional studies hadhad less schooling and，therefore，scored lower on intelligence tests than the middle。agedand young adult samples．Their test scores had not declined but had always been lowerthan those of the younger adults with whom they were compared．So the earlier cross'sectional research had discovered a cohort effect，not a true developmental change.

Despite this important limitation，the cross-sectional comparison is still the designdevelopmentalists use most often．Why?Because it has the advantage of being quick andeasy：we can go out this year,sample individuals of different ages，and be done with it．Moreover,this design is likely to yield valid conclusions when there is little reason to believe that the cohorts being studied have had widely different experiences while growingup．So if we compared 4．t05-year．olds with 7．t08-year-olds，as Coates and Hartup did，we might feel reasonably confident that history or the prevailing culture had notchanged in any major way in the 3 years that separate these two cohorts·It IS mainly instudies that attempt to make inferences about development over a span of many yearsthat cohort effects present a serious problem．

There is a second noteworthy limitation of the cross-sectional design：It tells us nothingabout the development of individuals because each person is observed at only one pointin time．So cross．sectional comparisons cannot provide answers to questions such asWhen will my child become more independent?orWill aggressive 2-year。olds be-come aggressive 5-year．olds?To address issues like these，investigators often turn to asecond kind of developmental comparison，the longitudinal design·

In a longitudinal design，the same participants are observed repeatedly over a period ottime．The time period may be relatively brief一6 months to a yea∞r it may be verylong，spanning a lifetime．Researchers may be studying one particular aspect of develop。ment．such as intelligence，or many．By repeatedly testing the same participants,investi'gators can assess the stability(continuity)of various attributes for each person in thesample．They can also identify normative developmental trends and processes by lookingfor commonalities，such as the point(s)at which most children undergo various changesand the experiences，if any,that children seem to share prior to reaching these mllestones．Finally,the tracking of several participants over time will help investigators to un-derstand individual differences in development，particularly if they are able to establishthat different kinds of earlier experiences lead to very different outcomes·

Several noteworthy longitudinal projects have followed children for decades andhave assessed many aspects of development(see KaganMoss，1962；Newman，Caspi，Moffitt．Silva．1997)．However,most longitudinal studies are much more modest in di。rection and scope．For example，Carolee Howes and Catherine Matheson(1992)con。ducted a study in which the pretend play activities of a group of l't02。yearolds wefereDeatedly observed at 6-month intervals over the next 3 years．Using a classificationscheme that assessed the cognitive complexity of play,Howes and Matheson sought todetermine fl)whether play did reliably become more complex with age，(2)whether chil'dren reliably differed in the complexity of their play,and(3)whether the complexity ofa child，s play reliably forecasted his or her social competencies with peers·Not surpns。ingly,all children displayed increases in the complexity of their play over the 3-year period although there were reliable individual differences in play complexity at each observation point．In addition，there was a clear relationship between the complexity ot achild's play and social competence with peers：Children who engaged in more complexforms of play at any given age were the ones who were rated as most outgoing and least aggressive at the next observation period 6 months later．So this longitudinal study showsthat complexity of pretend play not only increases with age but is also a reliable predic．tot of children's future social competencies with peers．

Although we have portrayed the longitudinal design in a very favorable manner,thisapproach has several potential drawbacks as well．For example，longitudinal projects canbe costly and time-consuming．These points are especially important in that the foCUS oftheory and research in the developmental sciences is constantly changing．and longitu．dinal questions that seem exciting at the beginning of a year project may seemrather trivial by the time the project ends．Practice effects can also threaten the validityof loElgitudinal studies：Participants who are repeatedly interviewed or tested may be．come test-wise or increasingly familiar with the content of the test itself，showing per．tormance improvements that are unrelated to normal patterns of development．Longitu．dinal researchers may also have a problem with selective attrition；children may moveaway or become bored with participating，or they may have parents wh0，for one reasonor another，will not allow them to continue in the study．The result is a smaller and p0．tentially nonrepresentative sample that not only provides less information about thedevelopmental issues in question but also may limit the conclusions of the study to thosechildren who do not move away and who remain cooperative over the long run．

There is another shortcoming of long．term longitudinal studies that students often seeright awav-the cross-generational problem．Children in a longitudinal project are typi．cally drawn from one cohort and are likely to have very different kinds of experiences thanyoungsters from other eras．Consider,for example，how the times have changed since the1930s and l940s when children in some of the early long．term Iongitudinal studies weregrowmg up．1n this age of dual-career families．more youngsters are attending day．care cen．ters and nursery schools than ever before．Modern families are smaller than those of yearspast，meaning that children now have fewer brothers and sisters．Families also move morefrequently than they did in the l930s and l940s，so many children from the modern eraare exposed to a wider variety of people and places than was typical in years gone by．Andno matter where they may be living，today's youngsters grow up in front of television setsand computers，influences that were not available during the l930s and l940s．So childrenof earlier eras lived in a very different world．and we cannot be certain that these young．sters developed in precisely the same way as today's children．Stated another wav．cross．generational changes in the environment may. Sequential designs combine the best features of cross-sectional and longitudinal studle5bv selecting participants of different ages and following each of these cohorts oVer tlme·To illustrate imagine that we wished to study the development of children's l091cal rea-soning abilities between the ages of 6 and l2．We might begin in 2002 by testing the l09ical reasoning of a sample of 6-year olds rthe l996 birth cohort)and a sample oi 8yeaolds fthe l994 birth cohort)．We could then retest the reasoning abilities of both group5in 2004 and 2006．Notice that the design calls for us to follow the l996 cohort from ages6 through lo and the l994 cohort from ages 8 through l2．A graphic representation ofthis research plan appears in Figure l．4．

There are three major strengths of this sequential design．First，it allows us to determin8whether cohort effects are influencing our results by comparing the logical reasonmg otsame．aged children who were born in different years．As shown in the figure，cohort efiectsare assessed by comparing the logical reasoning of the two samples at ages 8 and l0·If thesamples do not differ,we can assume that cohort effects are not operating·Figure l．4 also il-1ustrates a second major advantage of our sequential