Chapter 14 Higher-order regression analyses
14.1 Example data set
## Parsed with column specification:
## cols(
## VA = col_double(),
## IM = col_double(),
## EM = col_double()
## )
A data set on “motive congruence”.
A dataset containing the explicit intimacy motive, implicit affiliation/intimacy motive and self ratings of affective valence. The variables are as follows:
EM: Explicit intimacy motive, measured with a questionnaire adapted from the Unified Motive Scales, (Schönbrodt & Gerstenberg, 2012). Raw values have been \(z\) standardized.
IM: Implicit affiliation/intimacy motive, measured a with picture story exercise (6 pictures). Raw motive scores have been controlled for word count and \(z\) standardized.
VA: \(z\) standardized valence rating. Consists of two bipolar items from the PANAVA questionnaire (Schallberger, 2005): “zufrieden … unzufrieden” (satisfied … unsatisfied) and “ungluecklich … gluecklich” (unhappy … happy).
Each of the 362 rows is an individual participant.
14.2 Description
For the continuous tables, we present descriptive statistics.
| EM | IM | VA | |
|---|---|---|---|
| Mean | 0.00 | 0.00 | 0.00 |
| Std.Dev | 1.00 | 1.00 | 1.00 |
| Min | -3.57 | -1.98 | -3.04 |
| Median | 0.20 | -0.14 | 0.26 |
| Max | 2.65 | 5.54 | 1.49 |
| N.Valid | 362.00 | 362.00 | 362.00 |
| Pct.Valid | 100.00 | 100.00 | 100.00 |
14.3 Visualization
Figure 14.1: Relationships between key variables.
14.4 Regression analysis
| Model statistics | |
|---|---|
| Multiple \(R^2\) | 0.050 |
| Multiple \(R^2_{adj}\) | 0.037 |
| \(F_{5,356}\) | 3.755 |
| Sig. (\(p\)) | 0.003 |
| Resid. SD | 0.981 |
| Resid. df | 356 |
| Regression coefficients | |||||
|---|---|---|---|---|---|
| Coef. | Std. Err. | \(t\) stat. | Sig. (\(p\)) | VIF | |
| (Intercept) | 0.099 | 0.066 | 1.490 | 0.137 | |
| EM | −0.036 | 0.056 | −0.643 | 0.521 | 1.159 |
| IM | 0.106 | 0.061 | 1.742 | 0.082 | 1.398 |
| EMsqr | −0.089 | 0.033 | −2.655 | 0.008 | 1.154 |
| IMsqr | −0.014 | 0.025 | −0.558 | 0.577 | 1.485 |
| EM:IM | 0.146 | 0.051 | 2.874 | 0.004 | 1.077 |
14.4.1 Model checking
References
Schallberger, U. (2005). Kurzskala zur erfassung der positiven aktivierung, negativen aktivierungund valenz in experience sampling studien (panava-ks) [short scales for the assessment of positive affect, negative affect, and valence in experience sampling studies]. Retrieved from University of Zürich website: http://www.psychologie.uzh.ch/institut/angehoerige/emeriti/schallberger/schallberger-pub/PANAVA_05.pdf
Schönbrodt, F. D., & Gerstenberg, F. X. R. (2012). An IRT analysis of motive questionnaires: The unified motive scales. Journal of Research in Personality, 46, 725–742. Retrieved from http://www.sciencedirect.com/science/article/pii/S0092656612001353