Chapter 2 definitions Flashcards
Theory
- a general statement about the relationship between constructs/events.
- must be parsimonious
- must generate a testable hypothesis
Hypothesis
-formal prediction about the relationship between two or more variables which is logically derived from the theory
Independent Variable (aka treatment variable)
-determines how the groups in the experiment are divided
Dependent Variable
-(aka outcome variable or variable to be measured) is measured by the investigator and is used to compare the experimental groups.
Interaction
-how one independent variable affects the dependent variable depends on the other independent variable.
Manipulated Independent Variable
- investigator begins with a large number of participants and assigns them to experimental groups
Nonmanipulated Independent Variable (aka subjects variable)
Exists without the researcher’s intervention (ie. Participants are separated according to Character or physical traits, birth order etc.)
- participants are not randomly assigned
- problem with nonmanipulated variable is that it cannot be assumed that the participants are identical on average in the beginning because they are not randomly assigned
- cause/effect relationship difficult to determine in nonmanipulated variables
3 elements of an experiment
1) generate a theory
2) make a hypothesis
3) collect data to support hypothesis
Replication
-examine participant populations different from those in the original research so that it can be determined if these results can be generalized to a larger population
File Drawer Problem
-often only “significant results” are published so we do not know where we went wrong on non-significant experiments
Case Study Method
- in-depth evaluation of a single individual (or sometimes a few individuals)
- often involves a psychotherapy client suffering from a problem that the researcher is interested in
- ○ Client’s history, current behavior and changes in behavior over the investigation are recorded
- often involves descriptive data
limitations of case study (3)
- cannot generalize results from a single person to a large population
- issues determining cause/effect relationships
- investigators subjective judgements can influence results
Pros of case study method (2)
- appropriate when investigating a rare case (ie. multiple personalities)
- Appropriate when it can be argued that the individual being studied is no different from people on the dimension of interest (ie. Studying participants with a severed corpus collapsum; it can be argued that these participants have a “normal” brain and that these results can be compared)
statistical significance
- if the difference was large enough that in all likelihood it was not caused by chance
- traditionally a 0.5 significance level is used (ie. difference between scores is so large that it would occur less than 5% of the time by chance)
Correlation coefficient
- statistical test used to understand the relationship between two measures
- Need to examine reliability and validity before using any standardized test.
test-retest reliability coefficient
-researchers administer the test to a large number of people; this test is then re-administered to the same people some time later. The scores from the first administration are correlated with those from the second. A high positive correlation coefficient (ie. 0.90) indicates good consistency over time.
internal consistency
-Test is internally consistent when all the items on the test measure the same thing; high internal consistency coefficient can indicate that most items are measuring the same concept.
Face Validity
-do the test items appear to measure what they are supposed to measure.
Congruent validity (convergent validity)
-extent to which scores from the test correlate with other measures of the same construct; if two tests measure the same thing then the scores should highly correlate
Discriminant validity
-refers to the extent to which a test score does not correlate with the scores of theoretically unrelated measures.
Behavioral validation
-it is important that test scores predict relevant behavior as participants may behaviorally respond as they think they would act/how they wish they would act.