Psychometric properties of the Dickman Impulsivity Inventory

Transcription

1 Personality and Individual Di erences 29 (2000) 27±35 Psychometric properties of the Dickman Impulsivity Inventory L. Claes*, H. Vertommen, N. Braspenning Department of Psychology, Katholieke Universiteit Leuven, Tiensestraat 102, 3000 Leuven, Belgium Received 23 March 1999; received in revised form 1 June 1999; accepted 13 July 1999 Abstract The Dickman Impulsivity Inventory (DII) is a self report measure that distinguishes two types of impulsivity. Dysfunctional impulsivity is the tendency to act with less forethought than most people of equal ability when this tendency is a source of di culty. Functional impulsivity, in contrast, is the tendency to act with relatively little forethought when such a style is optimal. The current study presents an exploratory factor analysis of the Dutch DII and describes the psychometric properties of the instrument. It further clari es the nature of functional and dysfunctional impulsivity by examining the relation between these two traits and another self-report measure of impulsivity, namely the Eysenck Impulsiveness Questionnare (I 7 ). Finally, we also provide data concerning the relationship with demographics Elsevier Science Ltd. All rights reserved. Keywords: Functional impulsivity; Dysfunctional impulsivity 1. Introduction Impulsivity has been described by a number of overlapping and sometimes contradictory de nitions (McCown, Johnson & Shure, 1993; Webster & Jackson, 1997). These include ``human behavior without adequate thought'' (Smith, 1952), ``behavior with no thought whatsoever'' (English, 1928), ``action of instinct without recourse to ego restraint'' (Demont, 1933) and ``swift action of mind without forethought of conscious judgement'' * Corresponding author. Tel.: ; fax: address: laurence.claes@psy.kuleuven.ac.be (L. Claes) /00/$ - see front matter Elsevier Science Ltd. All rights reserved. PII: S (99)

2 28 L. Claes et al. / Personality and Individual Di erences 29 (2000) 27±35 (Hinslie & Shatzky, 1940). Impulsivity may also mean acting with minimal thought regarding future actions, or acting on thoughts that are not in the individual's or others' best interest (Anon, 1951). Attempts to describe the construct of impulsivity empirically are at least 40 years old and continue to this day. In the excellent book by Doob (1990), impulsiveness is de ned as the absence of re ection between an environmental stimulus and an individual's response. According to this de nition, the appropriateness of impulsiveness is largely a function of the demands of the situation at hand. Some stimuli, such as a car suddenly braking on a crowded freeway, may demand immediate behavior without deliberation. Lengthy re ection regarding all of the possible options may prove harmful or even fatal. On the other hand, many situations require careful re ection of all possible responses (McCown et al., 1993). Recent work on the relationship between impulsivity and cognitive functioning, also suggest that the consequences of impulsivity are not always negative. For example, when the experimental task is very simple, high impulsives, rapid responding has little cost in errors (Dickman, 1985). And when the time available for making a decision is extremely brief, high impulsives are actually more accurate than low impulsives (Dickman & Meyer, 1988). One question raised by these ndings (Dickman, 1990) is whether the factor that causes people to respond quickly and inaccurately when this style of responding is a source of di culty (dysfunctional impulsivity) is the same factor that causes people to respond quickly and inaccurately when this style is optimal (functional impulsivity). It could be that there is a general tendency to respond quickly and inaccurately that is sometimes a source of di culty and sometimes bene cial. Or, it could be that there are two separate traits, one that results in rapid inaccurate performance in situations where this is optimal and the other that results in rapid, inaccurate performance in situations where this is nonoptimal. To determine whether it was possible to discriminate within the self-report domain between functional and dysfunctional impulsivity, Dickman (1990) wrote a large number of items designed speci cally to tap these two types of impulsivity and then factor-analysed subjects' responses to these items in order to see whether the two hypothesised factors would emerge. The item pool used in this factor analysis contained 17 items written to tap functional impulsivity and 23 items designed to tap dysfunctional impulsivity. There were also 23 ller items. The rst two factors that emerged from this analysis clearly represented the two hypothesised components of impulsivity. The 11 items that loaded over 0.30 on factor 1 had all been written to tap functional impulsivity, whereas the 12 items that loaded over 0.30 on factor 2 had all been written to tap dysfunctional impulsivity. These 23 items (11 functional and 12 dysfunctional items) formed the Dickman Impulsivity Inventory, short version (DII-short). The current research was carried out in an attempt to provide a Dutch translation of the DII-short and to explore the factor structure and scale properties of this questionnaire in a Dutch-speaking Belgian sample. We further examine the convergence between the DII and another well-known measurement of impulsivity, the Eysenck Impulsivity Questionnaire (Eysenck & Eysenck, 1978). Finally, we present data on the relationship with educational level, age and gender. Most studies (Eysenck, Pearson, Easting & Allsopp, 1985) have found a

3 L. Claes et al. / Personality and Individual Di erences 29 (2000) 27±35 29 decline in impulsivity with age and show that men are equal or more impulsive than women (Eysenck et al., 1985; Luengo, Carrillo-De-La-Pena & Otero, 1991). 2. Method 2.1. Subjects Ss were 315 adults, representative for the Flemish population, who were recruited by students in psychology. Each student was given a pro le of three persons he or she had to look for, specifying gender, age and educational level in order to guarantee a good cross-section of the population. There were 159 male (50.5%) and 156 female (49.5%) respondents. Five di erent age ranges were represented: 18±24 years (16.5%), 25±34 years (25.3%), 35±44 years (25.0%), 45±54 years (18.8%) and 55±65 years (14.0%). With respect to educational level, 9.2% only attended elementary school, 52.8% went to high school and 37.3% attended higher education. The majority of the respondents (66.1%) were married or living together with a partner, 30.4% were unmarried, 1.3% divorced and 1.9% widowed. The distribution of the Ss over the gender groups, the age groups and the educational levels corresponds to the distribution in the Flemish population Instruments Dickman impulsivity inventory-short version (DII-short) The DII-short is a self-report questionnaire developed to measure two types of impulsivity, namely Functional and Dysfunctional Impulsivity. It consists of 23 items to be answered with a true/false answer format. Eleven items were written to tap functional impulsivity and consists of items such as ``I would enjoy working at a job that required me to make a lot of splitsecond decisions'' and ``People have admired me because I can think quickly''. Another 12 items were designed to tap dysfunctional impulsivity and consists of items such as ``I often say and do things without considering the consequences'' and ``I often say whatever comes into my head without thinking rst'' Eysenck impulsiveness questionnaire (I 7 ) The I 7 is a self-report inventory that consists of 54 items to be answered with a yes/no format. The authors (Eysenck & Eysenck, 1978) concluded that the I 7 is an adequate measure of three factors, named Narrow Impulsiveness (N-Imp, 19 items), Venturesomeness (Vent, 16 items) and Empathy (Emp, 19 items). In our study we are only interested in the rst two subscales. Impulsivity is aligned with Psychoticism (P) and Venturesomeness correlates with Extraversion (E), and both clearly measure di erent components of Impulsivity. Narrow- Impulsivity is measured with items such as ``I often buy things on impulse'' and ``I often do things on the spur of the moment'', Venturesomeness consists of items such as ``I quite enjoy taking risks'' and ``I like diving o the highboard''.

4 30 L. Claes et al. / Personality and Individual Di erences 29 (2000) 27± Procedure The two scales have been translated into Dutch and retranslated into English by a native speaker of English. Di erences between the original versions and the retranslations were discussed in order to improve the quality of the Dutch translations. Four hundred subjects were recruited by students in psychology. All subjects were asked to participate on a voluntary basis and anonymity was guaranteed. They received a letter informing them of the aims and the procedure of the study and asking their willingness to participate. If informed consent was obtained they were asked to ll out the instruments at their homes. Three hundred sixteen completed forms were recollected by the students, so we had a response rate of 80%. 3. Results 3.1. Factor analysis of the DII-short The correlation matrix of the 23 items was factor-analysed using the principal axis method, with the communalities estimated iteratively. The criterion for factor extraction was an eigenvalue equal or greater than 1. The factor matrix was rotated using an oblique rotation (oblimin, with delta set to 0). The rst two factors explained 37.3% of the total variance (Table 1) and represented the two expected components of impulsivity. Ten of the 11 items that loaded over 0.35 on factor 1 had all been written by Dickman to tap dysfunctional impulsivity, whereas 10 of the 12 items that loaded on factor 2 had been designed to measure functional impulsivity. Taken together we can say that the results of the oblique factor solution yield a similar factor pattern to that found by Dickman. Further evidence on that topic has been gathered by means of a Procrustes congruence rotation. We found a great concordance between the results in both studies. Congruence coe cients are good for both subscales, namely 0.92 for Dysfunctional Impulsivity and 0.88 for Functional Impulsivity. The reason why items 4, 8 and 23 of the Dutch DII did not t the original structure will be explored in Section Scales characteristics Means, standard deviations, internal consistency coe cients and subscale intercorrelations are presented in Table 2. As revealed by Table 2, the reliability analysis indicates su cient internal consistency of both subscales and shows that the Cronbach's alpha coe cients of the Dutch study are almost the same as those of the American study, namely 0.76 for the Functional Impulsivity subscale (American study: 0.74) and 0.84 for the Dysfunctional Impulsivity subscale (American study: 0.85). Whereas in the American study, the correlation between the two subscales was slightly positive (r = 0.23), in our study the two scales seem to be independent (r= 0.02).

5 Table 1 Results of the factor analysis with oblique rotation on the Dutch and American DII a Items This study (9) I often say and do things without considering the consequences (7) I often make up my mind without taking the time to consider the situation from all angles (22) Before making any important decision, I carefully weigh the pros and cons (18) I often get into trouble because I don't think before I act (1) Often, I don't spend enough time thinking over a situation before I act (10) I frequently make appointments without thinking about whether I will be able to keep them (21) I will often say whatever comes into my head without thinking rst (13) I frequently buy things without thinking about whether or not I can really a ord them (17) Many times the plans I make don't work out because I haven't gone over them carefully enough in advance (14) I'm good at careful reasoning (8) I have often missed out on opportunities because I couldn't make up my mind fast enough (11) I am uncomfortable when I have to make up my mind rapidly (6) I would enjoy working at a job that required me to make a lot of split-second decisions (15) I like to take part in really fast-paced conversations, where you don't have much time to think before you speak (5) I am good at taking advantage of unexpected opportunities, where you have to do something immediately or lose your chance (2) I try to avoid activities where you have to act without much time to think rst (20) People have admired me because I can think quickly (19) Most of the time, I can put my thoughts into words very rapidly (12) I don't like to do things quickly, even when I am doing something that is not very di cult (4) I enjoy working out problems slowly and carefully (16) I like sports and games in which you have to choose your next move very quickly (23) I rarely get involved in projects without rst considering the potential problems (3) I don't like to make decisions quickly, even simple decisions, such as choosing what to wear, or what to have for dinner Variance accounted for (%) a Factor loadings above 0.35 are printed bold. Factor 1 Factor 2 Factor 2 Factor 1 L. Claes et al. / Personality and Individual Di erences 29 (2000) 27±35 31

6 32 L. Claes et al. / Personality and Individual Di erences 29 (2000) 27±35 Table 2 Number of items, range, means, standard deviations, internal consistency coe cients and subscale intercorrelations for the DII a,b Subscale N Range M S.D. 1 2 (1) Functional Impulsivity 10 0± (2) Dysfunctional Impulsivity 10 0± ± a Italicized values on the diagonal are internal consistency coe cients (Cronbach's alpha's). b The construction of the subscales was based on the results of the factor analysis of the data of the Flemish population Correlations with other impulsivity scales We further clari ed the nature of Functional and Dysfunctional Impulsivity by examining the relation between these two traits and two subscales `Narrow Impulsivity' and `Venturesomeness' of the Eysenck Impulsiveness Questionnare (I 7 ) (Table 3). We found the same results as Dickman. Both the Functional and the Dysfunctional Impulsivity scales correlated signi cantly with all of the other impulsivity scales (all ps < 0.05). However, the Dysfunctional Impulsivity scale correlated signi cantly more highly with the `Narrow Impulsivity Scale' (r = 0.71, p < 0.001) and the Functional Impulsivity scale showed the strongest correlation with the `Venturesomeness Scale', a measure of adventurousness and risk taking (r = 0.46, p < 0.001) Relationship to demographics The relationships between the DII subscales and the demographic variables, sex, age and educational level were explored by means of univariate analyses of variance (ANOVAs). We found a signi cant e ect of sex, which was a consequence of the fact that men reported signi cantly more functional (F(1, 299)=14.80, p < 0.001) and dysfunctional (F(1, 293)=8.56, Table 3 Correlations between the Eysenck and the Dickman Impulsivity Subscales in both studies Eysenck's Impulsivity Scales This study Dickman's study (1990) l a 2 b Narrow Impulsivity Scale of Eysenck's I Venturesomeness Scale of Eysenck's I a Dickman's Functional Impulsivity Subscale. b Dickman's Dysfunctional Impulsivity Subscale.

7 L. Claes et al. / Personality and Individual Di erences 29 (2000) 27±35 33 p < 0.01) impulsive behavior than women. There was also a signi cant e ect of educational level on both functional (F(4, 295)=4.21, p < 0.01) and dysfunctional (F(4, 289)=4.46, p < 0.01) impulsivity; participants who went to university were more (dys)functional impulsive than those who only attended elementary or high school. In contrast with previous research (Eysenck et al., 1985; Luengo et al., 1991), we found no signi cant e ect of age. 4. Discussion The current study presents a factor analysis of the Dutch DII. As expected, a two-factor solution emerged, each factor representing one of the two types of impulsivity, named Dysfunctional and Functional Impulsivity. Except for three items, all items are found to belong to the same subscales in both versions, which again indicates that the internal factor structure of the DII is consistent across samples (American people versus Flemish people). This concordance between both versions of the DII was also demonstrated after congruence rotation, which yields good congruence coe cients for both subscales. The reason why items 4, 8 and 23 of the Dutch DII did not t the original structure need to be explored. Inspection of Dickman's original factor structure shows that item 8, compared with the other functional impulsivity items, has the smallest loading on the Functional Impulsivity Scale and can not be seen as a very clear example of a functional impulsivity item. This can be an explanation why this item loaded on both subscales, and did not appear to be a good indicator of functional impulsivity in our Dutch sample. The same can be said of item 4. In comparison with the other dysfunctional impulsivity items, it shows the smallest factor loading on the original Dysfunctional Impulsivity Scale. This explanation, on the contrary, can not be held for the mis t of item 23, because it has a substantial loading on the original Dysfunctional Impulsivity Scale. One possible explanation, is the fact that this item tends to have the same content as some of the items of the Dysfunctional Impulsivity Scale. Item 23 (``I rarely get involved in projects without rst considering the consequences'') expresses the need to think before acting, and resembles the content of item 22 (``Before making any important decision, I carefully weigh the pros and cons''), which belongs to the Dysfunctional Impulsivity Scale. The fact that item 23 has also a meaningful loading (0.28) on the this scale, can con rm this hypothesis. The internal consistency of the Dysfunctional Impulsivity Scale proved to be good, whereas that of the Functional Impulsivity Scale turns out to be moderate. This may imply that Functional Impulsivity in this questionnaire refers to a more heterogeneous set of phenomena than Dysfunctional Impulsivity. As in the original study of Dickman (1990), intercorrelation between the two subscales is low, suggesting that they are measuring two separate traits and are bearing di erent relations to other personality traits. We found signi cant correlations between the scale Dysfunctional Impulsivity and Eysenck's Narrow-Impulsivity Scale, both referring to actions of doing and saying things without thinking that lead to negative consequences. Functional Impulsivity, on the other hand, shows a signi cant positive relationship with Eysenck's Venturesomeness Scale, a measure of adventurousness and risk taking. This latter scale, also shows positive correlations with two subscales of Zuckerman's Sensation Seeking Scale, called `Risk-taking' and `Thrill and Adventure-seeking'. As we found that both Functional and Dysfunctional Impulsivity are independent traits, Eysenck and

8 34 L. Claes et al. / Personality and Individual Di erences 29 (2000) 27±35 Eysenck (1978) concluded the same with respect to their Impulsiveness and Venturesomeness Scales, both are primary, independent personality factors. As Eysenck (1993; The Impulsive Client, p. 59) noted, most researchers have interpreted impulsivity in the dysfunctional sense. It appears that there is little disagreement with Dickman's dysfunctional impulsivity scale measuring impulsiveness in a `psychopathological' sense, since it is signi cantly correlated with narrow impulsiveness as measured by the I 7. What is not adequately addressed in the literature is ``What does functional impulsivity represent?''. Is functional impulsivity similar to what Baars (1988) discussed as ``automatizing voluntary attention''? Are fast reaction times while solving complex cognitive tasks related to functional impulsivity? According to Dickman (1990), functional impulsivity is the tendency to act with less forethought when such a style is optimal. We further know that there is a positive relation between reaction times and general intelligence (Dickman, 1990), so we can formulate the hypothesis that high functional impulsives are more intelligent than low functional impulsives. Indirect evidence in favour of this hypothesis is given by the fact that we found a signi cant association between functional impulsivity and Educational Level. People with a university degree seemed to be more (dys)functional impulsive than those who only attended elementary or high school. However, further research in this domain is necessary to nd out more about the relation between functional impulsivity and cognitive information processing e.g. intelligence. Dickman (1990) further mentioned that previous studies (Eysenck et al., 1985) have found a decline in impulsivity with age and have shown that men are equal or more impulsive than women (Eysenck et al., 1985). Our data con rmed the nding that males are more impulsive than females, but they did not corroborate the assertion that there is a decline in impulsivity with age. The general purposes of this study were to evaluate a Dutch translation of the DII and to clarify the nature of functional and dysfunctional impulsivity as de ned by Dickman. The results of the oblique factor solution yield a similar factor pattern to that found by Dickman. Only three of the 20 three items did not t the original factor solution. More research will be necessary to determine whether these di erences are based on di erences in the populations or whether more studies will be necessary to determine the factor structure of the Dickman Impulsivity Inventory. References Anon (1951). Oxford English dictionary. Oxford, UK: Oxford University Press. Baars, B. J. (1988). A cognitive theory of consciousness. Cambridge: Cambridge University Press. Dickman, S. (1985). Impulsivity and perception: individual di erences in the processing of the local and global dimensions of stimuli. Journal of Personality and Social Psychology, 48, 133±149. Dickman, S. (1990). Functional and dysfunctional impulsivity: personality and cognitive correlates. Journal of Personality and Social Psychology, 58, 95±102. Dickman, S., & Meyer, D. E. (1988). Impulsivity and speed-accuracy tradeo s in information processing. Journal of Personality and Social Psychology, 54, 274±290. Demont, L. (1933). A concise dictionary of psychiatry and medical psychology. Philadelphia: Lippincott. Doob, L. (1990). Hesitation: impulsivity and re ection. New York: Greenwood Press. English, H. (1928). A student's dictionary of psychological terms. Yellow Springs, OH: Antioch Press.

9 L. Claes et al. / Personality and Individual Di erences 29 (2000) 27±35 35 Eysenck, H. J. (1993). The nature of impulsivity. In W. G. McCown, J. L. Johnson, & M. B. Shure, The impulsive client. Theory, research and treatment (pp. 57±69). Washington (DC): American Psychological Association. Eysenck, S. B. G., & Eysenck, H. J. (1978). Impulsiveness and venturesomeness: their position in a dimensional system of personality description. Psychological Reports, 43, 1247±1253. Eysenck, S. B. G., Pearson, P. R., Easting, G., & Allsopp, J. F. (1985). Age norms for impulsiveness, venturesomenes and empathy in adults. Personality and Individual Di erences, 6, 613±619. Hinslie, L., & Shatzky, J. (1940). Psychiatric dictionary. New York: Oxford University Press. Luengo, M. A., Carrillo-De-La-Pena, M. T., & Otero, J. M. (1991). The components of impulsiveness: a comparison of the I.7 Impulsiveness Questionnaire and the Barratt Impulsiveness Scale. Personality and Individual Di erences, 12, 657±667. McCown, W. G., Johnson, J. L., & Shure, M. B. (1993). The impulsive client: theory, research, and treatment. Washington, DC: American Psychological Association. Smith, L. (1952). A dictionary of psychiatry for the layman. London: Maxwell. Webster, C. D., & Jackson, M. A. (1997). Impulsivity: theory, assessment and treatment. New York/London: The Guilford Press.