Peter Schwenkmezger

CORTISOL AND BEHAVIOR: 2. APPLICATION OF A LATENT STATE-TRAIT MODEL TO SALIVARY CORTISOL

In two studies we applied a new stochastic model for the measurement of latent states and traits to salivary cortisol in healthy subjects. Data were analyzed by Linear Structural Relations (LISREL). In the first study, 48 students collected saliva samples on two occasions at 1300h with an interval of six weeks between occasions. Salivary cortisol levels were found to be mainly influenced by situational and/or interactional effects, which determined approximately 75% of the total variance of the hormone data, while only approximately 21% were determined by the latent trait. Measurement unreliability was found to be approximately 4%. In the second study, 54 young mothers collected saliva at 0800h, 1500h, and 2000h for three consecutive days, starting on the second day after delivery of a healthy baby. Early morning salivary cortisol showed high consistency across the three days. Approximately 60% of the total variance in the early morning cortisol samples was determined by the latent trait. For the 1500h and 2000h samples, however, the corresponding proportions of variances amounted only to 33% and 28%, respectively. These results appear to reflect the absence of a clear-cut circadian rhythm in some of the women on the second day after delivery.

Situational effects in trait assessment: The FPI, NEOFFI, and EPI questionnaires

While most researchers do agree now that situations may have an effect in the assessment of traits, the consequences have been neglected, so far: if situations affect the assessment of traits we have to take this fact into account in studies on reliability and validity of measurement instruments and their application. In the theoretical part of this article we provide a more formal exposition of this point, introducing the basic concepts of latent state–trait (LST) theory. LST theory and the associated models allow for the estimation of the situational impact on trait measures in non‐experimental, correlational studies. In the empirical part, LST theory is applied to three well known trait questionnaires: the Freiburg Personality Inventory, the NEO Five‐Factor Inventory and the Eysenck Personality Inventory. It is shown that significant proportions of the variances of the scales of these questionnaires are due to situational effects. The following consequences of this finding are discussed, (i) Instead of the reliability coefficient, the proportion of variance due to the latent trait, the consistency coefficient, should be used for the estimation of confidence intervals for trait scores, (ii) To reduce the situational effects on trait estimates it may be useful to base such an estimate on several occasions, i.e., to aggregate data across occasions. (iii) Reliability and validity studies should not only be based on a sample of persons representative of those to whom the test will be applied; they should also be conducted in situational contexts representative of the intended applications.

Do you feel better or worse? The validity of perceived deviations of mood states from mood traits

The advantages of assessing self‐perceived deviations of mood states from mood traits in cross‐sectional as well as longitudinal studies are discussed in detail. The validity of self‐perceived mood deviations is analysed in a longitudinal study with three occasions of measurement and 176 participants. In order to assess self‐perceived mood deviations, participants rated the deviation of their momentary mood from how they felt in general with respect to 32 mood adjectives belonging to three mood scales (pleasant–unpleasant, calm–restless, awake–sleepy). Furthermore, mood states, mood traits, and other personality variables (hedonic level, neuroticism, extraversion, openness, agreeableness, conscientiousness) are assessed by self‐report. Using models of latent state–trait theory, it is shown that the self‐perceived mood deviation scales are reliable (reliabilities between 0.73 and 0.95) and sensitive to occasion‐specific fluctuations (occasion specificities between 0.38 and 0.72). The self‐perceived deviation scales show high correlations with latent occasion‐specific deviation variables defined on the basis of repeatedly measured mood states (high convergent validity). In contrast to self‐reported mood states, however, self‐perceived mood deviations show much smaller and in most cases non‐significant correlations with personality variables (high discriminant validity). Furthermore, it is shown that self‐perceived mood deviations can be used to suppress stable situation‐specific variance in mood traits defined as aggregated states. Therefore, it can be demonstrated that including self‐perceived mood deviations in analyses on mood–personality associations enhances the association coefficients considerably. Finally, the implications of the results for the individual assessment of mood deviations, as well as for studies on affect and personality, are discussed. Copyright

A latent state-trait anxiety model and its application to determine consistency and specificity coefficients

Abstract In the present study the consistency model (Steyer, 1987) was applied to data gathered with the German version of the State-Trait Anxiety Inventory (Laux, Glanzmann, Schaffner, and Spielberger, 1981). The questionnaire was presented twice to 64 university students with an interval of two months between first and second testing. The consistency and specificity coefficients, estimated by LISREL (Joreskog and Sorbom, 1984), support the state-trait distinction. The state variables have high specificity and consistency coefficients; the trait variables, in contrast, have high consistency coefficients but low or even negligible specificity coefficients. The discussion points out the advantages of the consistency model over the stability model; the most important advantage is that the theoretical structure of the consistency model is more appropriate for the type of application considered. It contains a state factor for each occasion of measurement and a trait factor common to all occasions of measureme...

The emotional and cognitive components of trait anxiety: A latent state-trait model

Abstract A cross-validation of the latent state-trait anxiety (LSTA) model proposed by Steyer, Majcen, Schwenkmezger & Buchner (Anxiety Research, 1, 281–299, 1989) for test halves created from items of the German version of the State-Trait Anxiety Inventory (Laux, Glanzmann, Schaffner & Spielberger, 1981) fails. It is argued that the reasons for this failure are the heterogeneity of the items of the trait anxiety inventory and the fact that the trait inventory contains more ‘cognitive’ items than the state inventory. These arguments are supported by further analyses of the cross-validation data. If the trait test halves are replaced by new test halves which are constructed only from ‘emotional’ items, the LSTA model does not have to be rejected. If, however, the trait test halves are constructed only from ‘cognitive’ items, the model again has to be rejected. Theoretical parsimony is suggested as a general methodological principle. This principle is realized in the latent state-trait anxiety model because, instead of assuming different latent traits for the state and trait test halves, this model postulates only one single latent trait which conjoins the two different kinds of manifest variables.

Validating scales for the assessment of mood level and variability by latent state-trait analyses

Results concerning the validity of the mood survey, a questionnaire assessing two enduring aspects of mood—mood level and mood reactivity—are reported. The analyses show that both subscales are suitable for the assessment of relatively stable aspects of mood: sutuational and/or interactional effects only determine 3–11% of the variances of the scale scores, whereas 77–89% are due to true individual differences (not determined by situational and/or interactional effects). Furthermore, in order to examine to which degree the mood level scale assesses the mood level of a person averaged across several occasions, the correlation between the latent trait variable underlying the mood level scale and a latent trait variable defined as the expectation of repeatedly measured mood states was estimated. The correlation between these latent trait variables of r=0.78 indicates that the mood level scale is useful for the assessment of the average mood level. Finally, to validate the reactivity scale, subgroups differing with regard to the scores on the reactivity scale were analysed to examine the degree to which their actual mood is influenced by situational and/or interactional effects. The results show that the actual mood of people with high mood reactivity scores is influenced to a much higher degree by situational and/or interactional effects that the momentary mood of people with low mood reactivity scores.