Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
38 Cards in this Set
- Front
- Back
Gelso, 1979 Bubble hypothesis |
Every research design has it's own flaws, we get a clearer picture using multiple designs |
|
What resource is a good guideline for reviewing manuscripts |
Journal of counseling psychology reviewer guidelines, 2013 |
|
Citation for a good hypothesis is a testable research question that provides direction for an experimental inquiry |
Heppner, Wampold & Kivlighan, 2008 |
|
Citation for hypotheses should be phrased as falsifiable statements so they can be tested |
Popper, 1959 |
|
Research designs are often a trade off between what two tapes of validity |
Heppner, Wampold & Kivlighan, 2008 |
|
What is an error that often occurs in reporting reliability and validity |
Authors don't reference the sample from which estimates are deprived See Wilkinson, L. And the task force on statistical inference APA board of scientific affairs, 1999 |
|
What are some reliability statistics |
Heppner, Wampold & Kivlighan, 2008 Cronbach's alpha, KR-20, intraclass correlation coefficients, kappa, test-reste reliability |
|
7 types of validity |
1. Face validity 2. Content validity 3. Construct validity 4. Predictive validity 5. Concurrent validity 6. Convergent validity 7. Discriminant validity |
|
4 types of studies that vary in their different levels of internal and external validity |
Gelso, 1979 Experimenal field (moderate I, moderate e) Descriptive field ( low I, high e) Experimental laboratory (high I, low e) Descriptive laboratory (low i, low e) |
|
What is MAXMINCON |
Kerlinger, 1986 Maximize variance of experimental variables Minimize error variance Control for confound variables |
|
Cons of MAXMINCON |
May not apply to applied settings where perfect control is not possible |
|
What are methods of statistical control |
Multiple regression ANCOVA Partial correlations Residualizing |
|
What are disadvantages to statistical control compared to experimental control |
1. Assumes linear relationship between confound variables and outcome variables 2. Statistically controls for a "measured" confound variable, so it is possible that aspects of the confound are not accounted for my measurement 3. Cannot definitely rule out confounds and therefore attribution of causality is attenuated 4. May not be a me to account for problems due to not being able to randomly assign participants 5. Colinearity can be a problem if the confound variable correlates with predictor variables |
|
What are citatioms for describing the differences between quantitative and qualitative research |
Johnson & Christensen, 2008 Lichtman, 2006 |
|
Qualitative research assumes there are multiple realities, as many as there are participants. What field does this come from and what ontology does this represent? |
Anthropology Interpretivist-constructivist relativist (Guba & Lincoln, 1994) |
|
What aspect of people does qualitative research examine |
Experiential life of people Polkinghorne, 2005 |
|
What is the definition of quantitative psychology research? |
Cresswell, 2009 Psychological research that performs mathematical modeling and statistical estimation or statistical inference or a means for testing objective theories by examining the relationship between variables |
|
What sort of ontology does quantitative psychology research utilize |
Guba & Lincoln, 1994 Modified objectivity epistemology, viewing objectivity as an ideal |
|
What is it called when you combine both qualitative and quantitative research |
Hanson et al., 2005 Mixed methods |
|
What are advantages of a brief form measure |
Participants may be more likely to complete (Robins et al., 2002) Eliminate item redundancy (Robins et al., 2001) Long precedent for single item measures for things such as subjective well being, cultural identity, relationship intimacy, intelligence, self-esteem (Gosling et al., 2003) Reduce test fatuige |
|
Disadvantages of brief form measures |
At best may be a "reasonable proxy" (Gosling et al., 2003) Not accurate, not reliably, validity suspect |
|
10 confounds and 10 methods for controlling them |
1. Non equivalent groups - random assignment 2. Different histories - equivalent control groups 3. Maturation effect - equivalent control group 4. Testing effects - post test only design 5. Regression to mean - equivalent control 6. Instrumentation - equivalent control 7. Attrition - monitor for differential loss between groups 8. Diffusion of tx - test all participants once, use informed consent 9. Experimentor/participant effects - single/double blind experiment 10. Floor/ceiling effects - choose reliable/valid measures |
|
Definition of independent samples t test |
Inferential statistical test, determines whether there is a s.s. difference between the means in two unrelated groups Hypothesis test where we compare data from one sample to a population for which we know the mean but not the standard deviation |
|
When is independent samples t test used |
Posttest comparison between two independent groups Manipulation check Experiment when pretest data or other covariates are unavailable Post test differences between relevant outcome variables Test simple main effects between two time points |
|
Paired samples t test defenition |
Heppner et al Dependent samples t test compares means of two related groups to detect whether there are any s.s. differences between the means |
|
When is paired samples t test used |
Compare two means for a within groups design in which every participant is in both samples Subject differences across two time points - has a single sample changed from pre to post test Simple main effects across two time points |
|
Benefits of paired samples t test |
Based on the assumption that particular extraneous variables is important to the outcome of the study Accomplishes the same purpose of analysis of covariance it reduces the unexplained variance and yields a more powerful test |
|
Chi square test defentition |
Kerlinger & Lee, 2000 Tells us whether the results of a cross tabulation are statistically significant Are two categorical variables independent (unrelated) to one another |
|
When do you use chi square test |
Compare what we observe with what we expect Are there differences between independent groups in an outcome variable that is also categorical Will be significant if the residuals for one level of a varia me differ as a function of another variable |
|
Example of chi square use |
A manufacturer or watches takes a sample of 200 people. Each person is classified based in age and watch type preference (digital vs analog). The question is whether there is a relation between age and watch type preference IV/DV: age/preference |
|
ANCOVA defenition |
Analysis of covariance - type of ANOVA that is used to control for potential confounding variables. General linear model with a continuous outcome variable and two or more predictor variables where at least one is continuous and one is categorical |
|
Use of ANCOVA |
Test whether certain factors have an effect on the outcome variable after removing the variance for which covariates account |
|
ANCOVA example |
Examining the effects of ethnicity on a set of job related outcomes including attitudes towards co-workers, attitudes towards supervisor, feelings of belonging in the work environment and identification with the corporate culture when controlling for the sex of participants IV/DV: sexðnicity/work related outcimes |
|
Two way ANOVA defenition |
Factorial ANOVA Extension of a one way analysis of variance. There are two independent variables. Has 3 null hypotheses, three alternative hypotheses and three answers to the research questions Answer to research questions are similar to those provided for one way ANOVA but there are 3 |
|
When to use two way ANOVA |
Test for main effects and interactions between two independent variables |
|
Example of two way ANOVA |
Is there a s.s. difference in mean GPA by class level (freshman, sophomore, junior) and sex (m/f) Hypotheses: There is a s.s. difference in mean GPA by class level There is a s.s. difference in mean. GPA by sex There is a sig. Interaction between class me am and sex when predicting mean gpa |
|
Multiple linear regressing defenition |
An extension of a simple correlation. One or more variables are used to predict the outcome (criterion) Statistical method for studying separate and collective contributions of one or more predictor variables to the variation of the dependent vatiable. |
|
How is multipke linear regression used |
Describe how multiple predictor variables are related to a single criterion variable Can also predict magnitude and directionality of numerous variables effects on a single outcome variable by fitting a linear equation to the observed data |