Validity
What is differential attrition? (A potential threat to internal validity)
When participants who share common characteristics (or those from the same group) leave the study
What is between-subjects design (aka independent design)?
It compares scores from different groups of individuals (scores are considered independent because scores are provided by separate, unique participants)
What are the two main types of statistics?
Descriptive statistics and inferential statistics
Name the four elements of the experimental research strategy.
Manipulation, measurement, comparison, control
T/F: Causality can be inferred from non- and quasi-experimental designs.
False! NE and QE designs always contain a confounding variable or another threat to internal validity, so causality cannot be inferred.
What is statistical regression? (A potential threat to internal validity)
What is within-subjects design (aka dependent design)?
It compares scores in two or more different treatment conditions by observing or measuring the same individuals in each condition - so each individual in the study has more than one score
What is an independent variable? What is a dependent variable?
Independent: The variable being manipulated; this variable does not depend on other variables
Dependent: The variable that is observed for changes to assess effects of manipulation of the independent variable; does depend on other variables
Which type of design is susceptible to assignment bias, experimental design or non-/quasi-experimental design?
non-/quasi-experimental design (because the groups aren't randomly assigned/controlled)
What action may be taken to help mitigate the third-variable problem and the directionality problem?
Manipulation of variables
Name three threats to internal validity for within-subjects designs (100 points each!)
Environmental variables (i.e. weather changing); maturation; testing effects (practice, fatigue, carryover); statistical regression (effects of independent variables on dependent ones)
What are two advantages of between-subjects designs?
Easier to gather more participants; scores not contaminated by other treatment factors (fatigue, boredom, etc.); it's a design that can be used to answer many research questions (it may not always be the best option, but it CAN always at least be used)
What is the difference between a statistic and a parameter?
Statistic: a summary value describing a sample
Parameter: a summary value describing a population
One of the problems with causation is the third-variable problem. What is it?
The third-variable problem is the phenomenon of change(s) in one variable being accompanied by changes in many variables (e.g. increased food intake --> weight gain). A third variable may be controlling the way two variables are acting (e.g. stress may cause increased food intake and therefore, weight gain).
If a study has a p value of p < 0.07, is it statistically significant?
Yes! (Because P > 0.05)
Name three threats to internal validity for between-subjects designs (100 points each!)
Maturation, instrumentation (surveys, tests, interviews, other assessment materials), fatigue, statistical regression (the effects of independent variables on dependent variables)
What are two advantages of within-subjects designs?
Requires fewer participants than between-subjects designs; time-based data can be collected; because the participants are the same across conditions, they serve as their own baseline
We summarize descriptive statistics in part using central tendencies. What are three measures of central tendency?
Mean (the average score in the distribution), median (divides distribution in half and uses middle score) and mode (most frequently-occurring score)
Developmental designs are used to examine changes in behavior related to ______.
age
95%
What are two ways we can increase external validity even with strict controls on extraneous variables?
Simulation: creation of conditions that simulate or duplicate the natural environment
Field studies: research conducted in a situation that the participant considers a naturalistic setting
Name at least one disadvantage each for between- and within-subjects designs.
Between: Differential attrition (if participants who share common characteristics or are from the same group leave the study); communication between groups may influence one or both groups; individual differences between participants can become confounding variables if treatment groups differ systematically because of the differences
Within: Attrition; exertion; susceptible to time-related factors such as fatigue or environmental changes between conditions
What is variability, and what are two ways that we measure it?
Variability: The spread of scores in a distribution We measure it using:
Variance: the average squared distance of scores from the mean
Standard deviation: The average distance of scores from the mean
In _______-________ ______, different groups of individuals at different ages are measured and compared (between-subjects design). In _________ ______, we measure a variable over time for the same group of individuals (also between-subjects design).
cross-sectional design; longitudinal design
If the effect size of a study was 0.6 according to Cohen's d, would that be considered a small, medium or large effect size?
Medium effect size