Assessing Implicit Motives using the Multi-Motive Grid

Table of Contents

Introduction

Empirical Analysis of the MMG
Objectivity
Reliability
Retest Reliability
Internal Consistency
Distortedness of the MMG
Construct Validity
Intercorrelations of the Motive-components
Factorial Validity
Convergent and Divergent Validity
Criterion Validity

Discussion

References

Appendix: Checklist for Evaluating the MMG

Introduction

The Multi-Motive Grid (MMG) is a new instrument to measure implicit motives (Sokolowski, Schmalt, Langens & Puca, 2000). Schmalt (Schmalt, 1976) has developed the Grid-technique in order to combine the advantages of self-reports and projective tests (e.g. the TAT). Schmalt (1999) labeled these instruments “semi-projective”. Early semi-projective tests have measured all three implicit motives separately which made it necessary to use three different tests in order to assess all implicit motives. The MMG is an advancement of these single instruments and measures all three big implicit motives with their two components at once.

In difference to the TAT subjects do not have to write down a story but have to tick statements which are listed underneath each picture. Subjects have to decide whether statements fit to the depicted picture or not. The decision whether a motive-relevant statement fits to the picture or not, should be influenced by the implicit motives of the subject. Thus comparable to the TAT, it is expected that the pictures activate the implicit motive and one can conclude from the interpretation of the pictures on the implicit motives of the subject.

Also comparable to the TAT, subjects are confronted with ambigious pictures which often can be interpreted in an achievement-thematic, power-thematic and affliation-thematic way. In each statement, the subject have to decide whether the statement fits to the picture or not, although only the Yes-responses are considered in the calculation of the motive scores. The statements are considered to assess all three big implicit motives with both the Fear-component and the Hope-Component. Although the pictures can be interpreted in different ways, some pictures stimulate one motive more than the other. There are two pictures for each component which stimulate all three motives (polythematic), two pictures for each component which stimulate only two of the three motives (bithematic) and two pictures for each component which assess only one motive (monothematic). The MMG thus consists 14 pictures. The motive content of the pictures were ascertained by informing subjects about implicit motives and explicitly asking them (N=41) to rate the pictures on all three motives. Afterwards the mean of ratings for all pictures was calculated (Schmalt et. al, 1994).

Empirical Analysis of the MMG

The MMG fulfills the requirements of the Classical Test Theory. By including a closed response format and giving subjects only two response categories across statements, the conduction economy and evaluation economy can be increased, the reliability can be increased by having the opportunity to include more pictures, as well as the evaluation economy can be increased. In the following sections, I will briefly describe empirical findings regarding the quality criteria of the MMG.

Objectivity

In this section, I will briefly discuss the objectivity of the MMG and the manner how objectivity was achieved. The manual describes in depth how to conduct a testing situation with the MMG and which atmosphere of testing situation should be created. These instructions secure standardization of administration and standardization of the testing situation which increases the conduction objectivity of the MMG.

Furthermore, it is in depth described how to analyze motive scores of the subjects and a template is given in the test materials. Both the instructions how to calculate motive scores and the given template increase the evaluation objectivity of the test. Thus it can be assumed that the inter-rater reliability of the MMG will be high or perfect if the scorer follows the instructions. The test manual further describes in depth how to use the template and how to read off motive scores, percentiles and t-scores.

In order to secure interpretation objectivity, the test manual gives an in depth description of the motives (affliation, power and achievement) and their motive components (e.g. FF). Furthermore, it describes in a very good manner conflicts between incompatible motive components, e.g. having a high hope for affliation (HA) and having a high fear for rejection (FR). These conflicts or “ambivalence conflicts” within a motive are described for all motives. Giving a table with norms separated for all motives and separated for men and women further establishes a high interpretation objectivity of the MMG. Thus, I believe that the MMG fulfills all types of objectivity in a satisfied amount.

Reliability

Retest Reliability

In order to assess the reliability coefficients of the MMG, the Retest-Reliability and Internal Consistency by using the Alpha-Coefficient has been calculated. To determine the Retest-Reliability, 97 subjects had to fill out the MMG a second time with a time distance of 40 minutes. Two conditions have been realized in order to assess memory effects. In one condition, subjects were instructed to fill out the MMG a second time, independend from their answers of the first time, in the other condition, subjects were explicitly instructed to fill out the MMG like their first time. Table 1 shows the Retest-Reliabilities for both conditions.

illustration not visible in this excerpt

Table 1: Excluding Memory Effects

The Retest-Reliability coefficients seem to be higher for the Standard Condition than for the Memory Condition. However this difference is not significant, showing according to the authors that the MMG is robust against memory effects. The Retest Reliability could be appreciable increased compared to projective tests and lies between rtt = .77 and rtt = .92 and is thus good to very good according to O*NET.

However, I believe that the interval between retesting was too short and the finding of higher retest reliabilities of the MMG should be treated with caution. A longer time interval of 3-6 weeks should be tested in order to be able to evaluate the Retest Reliability and would further undermine that personality traits have been measured (implicit motives) and not state variables (explicit motives). The Retest Reliabilities for the fear components are slightly smaller (rtt = .77 to rtt = .80) than for the hope components (rtt = .88 to rtt = .92) but are still good to very good.

[...]