Child Development 133 
Research Methods in Human Development

Sections 01 & 05
Hembree            Fall, 2012

 

Exam Guide #3

Check out the exam handout for more information about the third exam (11/28). Revised 11/26/12

Terms:
External validity, generalizability (Supplementary Reading) :
external validity
ecological validity
conceptual and exact replication and importance of replication in science
relative benefits of lab and field studies

types of generalization (e.g., population, setting, time of interest)
threats to external validity (based on methods and participants)
ways to improve external validity and why/when should we?

trade-off between internal and external validity

Analyzing Data (Ch. 11 & t-test and Chi-square handouts)
* you may bring copies of the t-test and chi-square handouts (links below) with you for use on exam (please make sure that your copies do not have notes not related to these tests). T-test tables of critical values will be provided. BRING A CALCULATOR

descriptive versus inferential statistics
statistical hypothesis testing and logic behind it (what does "significant" mean?)
null hypothesis/experimental hypothesis
reject vs. fail to reject (& why we don't accept the null)
t-test - be able to do problems like those on problem set
be able to INTERPRET t-test and draw correct conclusions
alpha and type I and Type II error
power and what increases power
interpreting non-significant results (e.g., power or error)
Chi-Square (be able to compute and interpret)
critical values/using tables (for t-test & Chi Square) 
when to use t-test, chi-square
*be able to do problems like those on problem set #2

Developmental and quasi-experimental Designs (Ch. 13, Supplementary Reading)
(non-experimental) pretest-posttest designs (e.g., case study) - and problems with these designs
Special problems associated with the study of development/change
developmental designs as quasi-experimental designs
Cross-sectional vs. longitudinal designs (and advantages and disadvantages of each)
Cohort and problems with cohort in developmental research (and how related to internal and external validity)

Advanced Experimental and Correlational Designs (Ch. 7, 10, 12) :
levels and variables
questions to consider in designing studies (e.g., How many variables/levels to test?)
two-group experimental design (experimental group/control group)
random assignment vs correlated assignment (matched or natural pairs, repeated measures)
Independent (between groups) vs. correlated/repeated measures (within groups) designs
Repeated measures designs - Advantages and disadvantages of repeated measures designs and when to use
Order effects, practice effects, carry-over effects
Ways to increase design complexity and advantages of doing so
Factorial designs
main effects vs. interaction effects (and their interpretation)
be able to draw and interpret 2X2 design results
advantages of factorial designs
use of F-test and ANOVA analyses for factorial designs
correlation coefficients/testing for significance - be able to interpret
coefficient of determination
regression analysis
/multiple regression
multiple correlation coefficient (R) and percent variance accounted for (R2) in regression analysis
outcome/predictor variables
factor analysis

**Note: be able to identify or evaluate a study's design

 

Short Essay

One of the following questions will be selected for the essay portion of the exam.

1) How is it that developmental psychologists go about studying development (or change)? Discuss the difficulties associated with studying development and the advantages and disadvantages of longitudinal and cross-sectional designs. What role does cohort play in these designs and how does cohort affect external or internal validity in these designs? (Be sure to provide examples.)

2) Discuss different ways that scientists increase the complexity of designs (e.g., increased # of levels of a variable, factorial designs) and the advantages and disadvantages of doing so. As an example, design a factorial study to test the effects of having a peanut butter sandwich and/or orange juice for lunch on children's test performance. If you conducted the study and found a main effect for having the peanut butter sandwich and an interaction between peanut butter and orange juice, what might your data look like (draw and label a graph and/or give hypothetical results in a table)? How is this design an improvement over separate studies examining the effects of peanut butter and orange juice?

 

Back to top

 

 

Send problems, comments or suggestions to: hembrees@csus.edu

California State University, Sacramento
College of Education
Department of Child Development

Updated: August, 2012