- The ItemsMethodResultsDiscussionReferences
Summary
648 subjects rated themselves on a German version of Underwood and Froming's "Mood Survey". The 15 item measure is a personality instrument for happy and sad moods with a clear-cut two-factor structure: Chronic Mood Level and Reactivity. The two subscales are internally consistent and moderately negatively correlated. Females had significantly higher values on both scales. Extending previous research, correlations between the mood subscales and the Big Five and several additional personality factors are reported. Mood Level is positively correlated with the personality factors Extraversion, Emotional Stability, Arousal/Happiness, and a one item measure of state pleasure. The personality correlates of the Reactivity subscale are less clear.Introduction
The Stimmungsskala is a German adaptation of Underwood and Froming's Mood Survey (Bohner, Hormuth and Schwarz, 1991; Underwood and Froming, 1980). The test has been developed to study the relationship between mood and personality. It comprises two subscales, namely Chronic Mood Level (sad, happy) and Reactivity (frequency and intensity of mood changes). Both scales are intended to measure enduring aspects of mood which may be regarded as personality traits.The German language items (and the original English language items) are:
(1) Manchmal pendelt meine Stimmung mehrmals zwischen glücklich und traurig in einer einzigen Woche.
I may change from happy to sad and back again several times in a single week.
(2) Ich fühle mich meist ziemlich fröhlich.
I usually feel quite cheerful.
(3) Meine Stimmung ist oft bedrückt.
I'm frequently "down in the dumps".
(4) Ich sehe im allgemeinen mehr die Sonnenseiten des Lebens.
I generally look at the sunny side of life.
(5) Verglichen mit meinen Freunden gehen meine Stimmungen weniger rauf und runter.
Compared to my friends, I'm less up and down in my mood states.
(6) Ich bin selten in wirklicher Hochstimmung.
I'm not often really elated.
(7) Manchmal schwankt meine Stimmung sehr schnell hin und her.
Sometimes my moods swing back and forth very rapidly.
(8) Ich fühle mich meist so, als ob ich vor Freude übersprudeln würde.
I usually feel as though I'm bubbling over with joy.
(9) Meine Stimmungen sind sehr konsistent; sie ändern sich fast nie.
My moods are quite consistent; they almost never vary.
(10) Ich halte mich für eine glückliche Person.
I consider myself a happy person.
(11) Verglichen mit meinen Freunden denke ich weniger optimistisch über das Leben.
Compared to my friends, I think less positively about life in general.
(12) Ich bin eine Person, die sich oft ändert.
I'm a very changeable person.
(13) Ich bin nicht so fröhlich wie die meisten Leute.
I am not as cheerful as most people.
(14) Ich bin weniger von meinen Stimmungen abhängig als die meisten Leute, die ich kenne.
I'm not as "moody" as most people I know.
(15) Meine Freunde scheinen oft zu glauben, daß ich unglücklich bin.
My friends often seem to feel I am unhappy.The scale values range from 1 ("trifft auf mich überhaupt nicht zu")
to 5 ("trifft auf mich ganz genau zu").Bohner et al. reported that the German adaptation has the same factorial structure as the English version. They found that the "subscales show a moderate negative inter-correlation and a high level of internal consistency and test-retest-reliability, over a five-week-period. The scores are independent of social desirability and are meaningfully correlated with personality measures of sociability and level of activation" (Bohner et al., 1991, p.135).
The present paper presents a replication of some findings of Underwood and Froming (1980) and Bohner et al. (1991). By relating the mood subscales to the Big Five and several additional personality factors which have not been regarded in previous research it provides some new information on mood-personality correlations.Method
Mood ratings were collected in three experiments comprising a total of 648 subjects (324 males and 324 females). Subjects' age ranged from 15 to 67 years (median: 23, interquartile: 21-26).
Prior to judging themselves on the Stimmungsskala, the subjects had taken part in an experiment on facial judgment. They had also rated themselves on 67 bipolar adjective scales tapping the Big Five, several facets of physical attractiveness, and some aspects of mood (arousal and happiness). Furthermore, the subjects had judged themselves on Russell's Affect Grid (Russell, Weiss and Mendelsohn, 1989). The Grid consists of 49 cells which are arranged as a square of 7 rows by 7 columns with the rows defining the present level of arousal and the columns defining the present level of pleasure. By choosing the appropriate cell the subject simultaneously reports both aspects of his or her affective state.Results
A principal component analysis of the Stimmungsskala yielded three eigenvalues greater than 1: 5.42, 2.51 and 1.04, accounting for 36.1, 16.7 and 6.9 per cent of the total variance, respectively. In accordance with Underwood and Froming (1980) and Bohner et al. (1991) only two components, henceforth referred to as factors, were extracted and rotated using Varimax.1 The loadings are listed in Table 1.
--: absolute value less than 0.30. Items 3, 5, 6, 9, 11, 13, 14, and 15 reflected.Tab. 1: Factor Loadings Item Mood Level Reactivity 2 .79 -- 10 .75 -- 4 .75 -- 6 .74 -- 8 .69 -- 13 .69 -- 11 .68 -- 3 .65 -.34 15 .48 -- 7 -- .83 9 -- .79 1 -- .78 5 -- .73 14 -- .64 12 -- .52 In accordance with previous findings the factorial structure is clear-cut. With the exception of item 3 no secondary loading exceeds 0.3 in absolute value. In the original version this item had a loading of the same magnitude on both factors (Underwood and Froming, 1980, study II); in the German version, however, it clearly belonged to the Level factor (0.66 vs. 0.32; Bohner et al., 1991).
Scale values were computed as unweighted means of the 9 Level and the 6 Reactivity items, respectively. The alpha reliabilities are 0.870 and 0.829. These values are slightly higher than Bohner et al.'s (0.86 and 0.78). Dropping item 15 would raise the internal consistency of the Level scale to 0.872; and deletion of item 12 would raise the reliability of the Reactivity scale to 0.843.
Table 2 shows the frequency distribution of the scale values.
Higher values indicate more positive mood and higher frequency and intensity of mood changes, respectively.Table 2: Frequency distribution of scale values. Percentile 10 25 50 75 90 Mean Std.dev. Mood Level 2.44 3.00 3.56 4.00 4.44 3.49 0.75 Reactivity 2.17 2.67 3.33 4.00 4.33 3.28 0.85 Females had significantly higher values on the Level scale (3.60 vs. 3.38; t=3.87, p<0.001), i.e., females reported to be happier than males. They also had significantly higher Reactivity values (3.41 vs. 3.15; t=3.89, p<0.001), i.e., females reported more frequent and more intensive mood changes. These sex differences are only partly consistent with previous findings. Both Underwood and Froming and Bohner et al. reported significantly higher Reactivity values for females but no significant difference on the Level subscale.
Replicating the findings of Underwood and Froming and Bohner et al. the two subscales were negatively correlated (r=-0.34; p<0.001). Controlling for sex of subject yields an inter-scale-correlation of -0.38 (p<0.001).
To check the concurrent validity of the Stimmungsskala the mood scales were correlated with several external personality measures. A principal component analysis of the 67 bipolar adjective scales yielded 8 factors accounting for 53.4 percent of the variance (Henss, 1996). Using Varimax rotation, 8 orthogonal factors were extracted and factor scores computed. The factors were straightforwardly interpretable as the Big Five factors Extraversion, Agreeableness, Conscientiousness, Emotional Stability and Intellect/Openness, as well as Physical Attractiveness, Health, and Arousal/Happiness. The left part of table 3 shows the correlations between the mood scales and the external personality factors. Also included are the correlations with present arousal and pleasure as measured by the Affect Grid, age, self reported height and weight, and the derived Body Mass Index (BMI = weight in kg / height in cm2).
Tab. 3: Mood-Personality Correlations. Simple Correlation Partial Correlation Mood Level Reactivity Mood Level Reactivity Extraversion .41 n.s. .43 .17 Agreeableness .14 -.20 n.s -.19 Conscientiousness -.09 -.13 -.17 -.19 Emotional Stability .38 -.40 .34 -.25 Intellect/Openness n.s. .14 .14 .18 Arousal/Happiness .46 -.23 .41 n.s. Physical Attractiveness .17 n.s. .16 .08 Health .12 n.s. .08 n.s. Grid Pleasure .46 -.21 .40 n.s. Grid Arousal n.s. n.s. n.s. n.s. Age n.s. -.18 -.10* -.19* Body Height -.15 -.11 n.s. n.s. Body Weight -.13 -.15 n.s. -.08 Body Mass Index n.s. n.s. n.s. n.s. n.s.: p>0.05; *not controlled for age of subject Positive mood as measured by the Level scale correlates positively with state pleasure as measured by the Affect Grid (0.46), and factor scores on Arousal/Happiness (0.46), Extraversion (0.41), and Emotional Stability (0.38).
Frequency and intensity of mood changes as measured by the Reactivity scale are negatively related to the Big Five factor Emotional Stability (-0.40).
Because Level and Reactivity are negatively correlated and because of the significant sex differences partial correlations were computed for each subscale while controlling for the other and sex of subject. Since both mood scales were significantly correlated with age, age of subject was also partialled out. The partial correlations are reported in the right part of table 3. Controlling for Reactivity and sex and age of subject has little effect on the Level-personality correlations. The positive correlations with Grid Pleasure, Extraversion, Arousal/Happiness and Emotional Stability remain of the same magnitude. On the other hand the correlation between Reactivity and Emotional Stability drops from -0.40 to -0.25 when the influence of Level, sex and age are removed.
For both mood scales there are several additional significant correlations. These, however, are rather low.Discussion
The Stimmungsskala has a clear-cut factorial structure. Its two subscales are easily interpretable as Chronic Mood Level and Reactivity. The scales are internally consistent and moderately negatively correlated. Consistent with previous findings females reported significantly more frequent and intensive mood changes as measured by the Reactivity subscale. In the present study, but not in previous ones, females also had significantly more positive values on the Level subscale. Both scales show some meaningful correlations with external personality factors and mood state measures. This is particularly true for the Level subscale. Positive mood is positively related to the personality factors Extraversion, Emotional Stability, Arousal/Happiness, and to state Pleasure as measured by Russell's Affect Grid. These correlations are independent of mood Reactivity and sex and age of subjects. The mood-personality correlations concerning the Reactivity subscale are somewhat problematic. Removing the influence of mood Level, age and sex reduces the correlation between Reactivity and Emotional Stability from -0.40 to -0.25. Thus all partial Reactivity-personality correlations are rather low. Underwood and Froming (1980) suggested that Reactivity correlates with temperament dimensions involving two sorts of "reaction to a stimulus situation: emotionality and impulsivity" and Level correlates "only with those temperaments which are not reactive" (p.410). This pattern is not corroborated by the present study. To clarify the personality correlates of mood Level and Reactivity further investigations are needed.
1The third factor of the three-factor solution is constituted by items 12 and 15. This factor would be difficult to interpret.References
- Bohner, G., Hormuth, S.E. and Schwarz, N. (1991). Die Stimmungs-Skala: Vorstellung und Validierung einer deutschen Version des "Mood Survey" (The Stimmungs-Skala: Presentation and validation of a German version of the "Mood Survey"). Diagnostica, 37, 135-148.
- Henss, R. (1996). The Big Five and physical attractiveness. Poster presented at the 8th European Conference on Personality. Ghent, Belgium, July 8-12, 1996.
- Russell, J. A., Weiss, A. and Mendelsohn, G. A. (1989). Affect Grid: A single-item scale of pleasure and arousal. Journal of Personality and Social Psychology, 57, 493-502.
- Underwood, B. and Froming, W. J. (1980). The Mood Survey: A personality measure of happy and sad moods. Journal of Personality Assessment, 44, 404-414.
