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;
62 Cards in this Set
- Front
- Back
Systematic Between Groups Variance |
differences between the groups means as a result of the manipulated variable |
|
Nonsystematic, Within Groups Variance |
Average variation within each group on the DV resulting from additional variables not controlled or manipulated in a study |
|
T-Test Evaluate? |
Does the difference between the means of 2 groups exceed what would be expected from sampling error or chance |
|
Examples of T-Test |
Independent Samples Dependent Samples |
|
Null Hypothesis |
All the group means are equal in the target population, any differences found in a sample is the result of chance fluctuations |
|
ANOVA Stats |
F= Ratio of between group variance to within group variance (X,Y)= Degrees of freedom between groups and within groups respectively P= Probably of some data if null hypothesis is true Etq Squared (h2)= Effect size statistic similar to r squared |
|
Post-Hoc Test |
Follow-up comparison tests that are conducted following a significant F test to determine which group differences are statistically significant, while controlling for elevated type 1 error rates |
|
Post-Hoc test examples |
Tukey and Bonferroni |
|
Anova-Stat. sign. F Ratio? Yes |
Reject Null and conclude there is a relationship between IV and DV--conduct additional analyses to compare all the group means while controlling for Type 1 |
|
Anova-Stat. sign. F ratio? No |
retain null and conclude insufficient data to rule out chance |
|
Axiom |
Clients behavior areinfluenced by multiple variables,not just a single independent variable! |
|
Interaction Effect
|
(The sum is greater than the 2 halves) -Effect of 1 IV on the DV depends on the level of a second IV -Effect of 1 IV is not consistent across all groups of 2nd IV |
|
ANCOVA |
Evaluates whether the differences among group means are likely due to chance or sampling error, after controlling for the influence of a covareist |
|
Quasi-Experimental Designs |
Identify casual relationships among research variables when random assignment is not feasible for ethical or practical reaseons |
|
Quasi-Experimental Design procedure |
Non-equivalent, pre-existing groups are compared on a DV after being exposed to different levels of the IV |
|
Major Drawback of Quasi-experimental Design |
Lack of random assignment reduces the internal validity of the studies results (selectional biases) |
|
Temporal Precedence |
-Cause must precede effect -Must be a functional relationship between A and B -Alternative explanations for functional relationship between A and B must be ruled out |
|
Confounding variable hypothesis |
Another factor other than the IV was responsible for changes in the DV (Major threat to internal validity) |
|
Problems with between subjects experiments |
Require large number of participants so the groups within have adequate numbers of participants May have low statistical power because of random error |
|
Within subjects design |
All participants are exposed to all levels of the IV, ideally in a random fashion |
|
Matching or dependent samples design |
Reduce error variance by intentionally matching participants confounding variables |
|
Experimental Designs-Between Groups |
Participants are separated into distinct groups and the average group scores are compared on a DV |
|
Experimental Designs-Within groups |
Every Participants experiences all levels of the IV: we compare the same people across all levels of IV |
|
Experimental Design-Matched Groups |
Matched pairs are identified and the members of these pairs are randomly assigned to different levels of the IV |
|
Basic Research
|
Researchdirected toward theory development and understanding the human condition. The discovery of new knowledge is valuable inand of itself.
|
|
Applied Research
|
Researchdirected toward solving practical, real-world problems
|
|
Outcome Research |
Evaluate the efficacy of a counseling model, set of interventions, or formal treatment program. |
|
Program Assessment and Evaluation |
Not doing research to benefit scientific community; rather for local stakeholders |
|
Program Assessment objective |
Identify the needs and priorities of a local population, organization, or school. |
|
Types of Program assessment and evalution |
Data-driven Perceptions based |
|
Data-driven Program assessment and evaluation |
Needs are identified through an evaluation of (relatively) objective data; these data become the foundation for progammatic changes. |
|
Perceptions-base program assesment |
Needs are identified through soliciting the needs and priorities of various stakeholders |
|
Program Evaluation Objective |
Evaluate the effectiveness and benefits of a specific program for a specific group of individuals |
|
Steps to Program evaluation |
Needs assessment Develop goals and objective Design and implement Program Data collection and Analysis Communicate Results to Stakeholders Use results and stakeholder evaluations to inform next steps |
|
Social Constructivism |
There is no single objective reality independent of an observer, reality is created in the mind of each individual and interpretations attached to his or her subjective experiences |
|
General Research Strategies of Qualitative Research |
Intensive interviewing Participant Observation Discovering themes, regularities, and categories in participants' personal stories/experiences Reflective Writing and observation Subjective interpretations of data Documentation of the researcher's biases and perspectives |
|
Ethnographic research objective |
Describe or explain the collective experiences and behaviors of a specific cultural group in its natural setting |
|
Grounded theory objective |
Develop a theory to account for a particular phenomenon. |
|
Phenomenological Research objective |
Describe the subjective experiences of a group of individuals who have a particular phenomenon |
|
Multiple Regression Stats |
R= Correlation between actual criterion scores and predicted criterion scores (team scores) R(Squared)= Proportion of variance in criterion variable explained by the combination of predictor variables b or B= Unique contribution of a predictor variable to the prediction of the criterion variable (Individual Points) |
|
Basic Survey Research Objective |
Identify and describe attitudes, opinions, behaviors, or other characteristics of a group of respondents |
|
Basic survey research, basic strategy |
Collect data through administration of a questionnaire or interview to ideally a representative sample of a population |
|
Basic Survey Research examples |
Needs assessment Epidemiology survey Cross-sectional and longitudinal research |
|
Ex Post Facto/ Casual Comparative Objective |
Evaluate whether pre-existing groups (IV) differ on select research variables (DV) |
|
Ex Post Facto Strategy |
Measure characteristics of 2 or more groups on one or more research variables |
|
Ex Post Facto Challenges |
Groups should, ideally, differ only on IV (Grouping Factor) |
|
Ex Post facto IV |
Categorical Group Membership |
|
Ex post facto DV |
Outcome variable measure by researcher |
|
Basic Format of Correlational Research |
Every participant has numerical scores/ratings on at least 2 research variables--an IV and DV |
|
Correlational Design Objective |
Explain relationship between 2 or more research variables Make predictions about how individuals will score on 2 variable from how they score on another variable |
|
Correlational Design Strategy |
Measure participants on 2 or more variable and use statistical procedures to describe relationship among variables |
|
Pearson Correlation Coefficient |
An index of the linear relationship between 2 continuous variables that indicates the extent to which changes in 1 variable correspond to changes in a 2nd variable |
|
Correlational Coefficient Range |
-1.00 to +1.00 |
|
Coefficient of determination (r Squared) |
Percentage of variance shared between 2 variables |
|
Interpreting effect size (Small) |
.10 |
|
Interpreting effect size (Medium)
|
.30 |
|
Interpreting effect size (Large) |
.50 |
|
Moderator variable |
A 3rd variable that changes the strength or the direction of the relationship between X and Y |
|
Mediator Variable |
3rd variable that accounts for, or at least substantially reduces the relationship between X and Y. Essentially, it explains why the relationship between X and Y occur. |
|
Steps in linear regression |
Find regression line that best fits the data points shared by 2 variables, minimizing errors of prediction. Slope intercept form: y= (X)b+a Use the equation to predict future scores on Y variable |
|
Multiple Regression Purpose |
Predict scores on a criterion variable (Y) from scores on 2 or more predictor variables (x1, x2) |
|
Multiple Regression Benefit |
Results indicate how well the predictor variable (X), alone and in combination explain variation in the criterion variable (Y) |