- are able to describe the differences between experimental, quasi-experimental, and correlational research designs, and they are able to infer from research examples which research design was used;
- are able to describe and reproduce in their own words the basic concepts of simple regression. These basic concepts include: linear association, least-squares estimation, explained variance, Multiple R-square, multiple correlation, adjusted R-square, raw and standardized regression coefficients, maximum direct effect, model tests, predicted scores, and residuals;
- are able to explain the differences between zero-order, partial, and semi-partial (part) associations, both from a statistical perspective and a substantive point of view; they are able to link the different associations with real research questions; they are able to reconstruct from a Venn diagram, the zero-order, partial, and semi-partial correlations and multiple R-square values for regression models with two continuous predictors;
- are able to execute a (logistic) linear multiple regression analyses (using SPSS) for models including continuous and categorical predictors; both for models with main effects only, and models including interactions between categorical and continuous variable, or two continuous variables;
- are able to correctly interpret practical applications of hierarchical regression analysis; are able to test significance of the R-square change using the F-test (using SPSS and by hand) and are able to correctly interpret the results; are able to design the hierarchical analysis given the research question envisaged;
- are able to choose the appropriate regression analysis given the research question envisaged;
- are able to draw the correct substantive conclusions from results of the regression analyses, within the context of psychological research;
- are able to reproduce the four basic assumptions of multiple regression analyses and they are able to test the tenability of the assumptions in real data;
- are able to apply the following concepts from inferential statistics (as taught in the course Introduction to Statistics) within the context of multiple (logistic) regression analysis: null hypothesis significance testing, one-tailed versus two-tailed testing, test statistics, p-values, confidence intervals, Type I and Type II errors, power, precision, statistical versus practical significance, effect sizes;
- know the APA guidelines with respect to reporting the results of (hierarchical) multiple (logistic) regression;
- know the commonly accepted rules of thumb for interpretation of effect sizes in terms of explained variances;
- know the concepts of probabilities, odds and logits; they know the relationship between the three scales; they are able to transform one into another (formulae are provided).
This course includes the following subjects:
- Correlational research designs
- Linear regression analysis: simple and multiple regression analysis, tests of single regression parameters, tests for model comparison, tests for a set of parameters, categorical predictors using dummies, interactions and probing of simple effects when significant interactions are found, model assumptions and testing their tenability in real data.
- Binary logistic regression analysis, using continuous and categorical predictors, including hierarchical logistic regression, and logistic regression with interaction effects.
- Tutorials. We offer tutorials in which the subjects are discussed and where students can practice the matters at hand. Although the tutorials are not compulsory, we do expect students to follow them. They are complementary to the lectures. We also expect active participation of the students in the tutorials. Students are not allowed to attend other tutorials than the one they did subscribe for.
- SPSS lab sessions. Part of this course are two lab sessions in which students have to complete exercises in SPSS. Attendance and active participation is compulsory. In case one or both lab session are missed, students should pass the SPSS test to fulfill the SPSS requirement. There are no other possibilities, or replacement assignments. The lab sessions need to be passed in order to pass this course. Students are expected to follow the lab session in the same year as they followed the course.
- Exams and final grade. There will be four lecture tests. These lecture tests are administered at the end of the tutorials. Students receive a grade for each lecture test. The final grade of the course is based on the grade for the exam and the average of the lecture tests, in which the lowest grade of the lecture test is crossed out (this means that students can miss one test without consequences). The final grade is obtained as follows: if the exam grade exceeds the average grade for the lecture tests, than the exam grade is also the final grade. If the exam grade is less than the average grade for the lecture tests, then the final grade is 0.8x exam grade + 0.2x average grade of the lecture tests. There is no separate resit for the lecture tests. Important! The grades for the lecture tests are only valid for the current academic year. The lecture grades never transfer to future or past academic years.
|Course available for exchange students|
|Written test opportunities|
|Written test opportunities (HIST)|
|Schriftelijk / Written||EXAM_01||BLOK 1||1||11-12-2018|
|Schriftelijk / Written||EXAM_01||BLOK 3||2||02-04-2019||Required materials|
|The book that is used in this course is a customized edition of the book Bivariate Through Multivariate Techniques by Rebecca Warner. This edition, specifically developed for TSB, is only available via the study association or via Studystore.|
|Title||:||Introduction to Techniques for Causal Analysis|
|Lecture notes (available on Blackboard)|