SEARCH

SEARCH BY CITATION

Keywords:

  • Barratt Impulsiveness Scale;
  • impulsivity;
  • reliability;
  • validity

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

Abstract No instrument for assessing impulsiveness has been developed in Japan. After translating the Barratt Impulsiveness Scale 11th version (BIS-11) into Japanese, we investigated reliability and validity in student (n = 34) and worker (n = 416) samples. To assess test–retest reliability, the intraclass coefficient between test and retest was calculated in the student sample. Internal consistency was examined by calculating Cronbach's alpha in the worker sample. To see factor validity, we examined by confirmatory factor analysis whether the three-factor model, proposed by a previous report, fit the data. The results showed that the Japanese version of the BIS-11 had excellent test–retest reliability and acceptable internal consistency reliability. In addition, the Japanese version was judged to have similar factor structure to the original one. The Japanese version of the BIS-11 is a reliable and valid measure and has possible utility for assessing impulsiveness.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

In recent years, much attention has been paid to impulsiveness of psychiatric patients by clinicians and researchers. In such situations, a number of useful instruments for assessing impulsiveness have been developed, and the literature on impulsiveness has increasingly grown.1–9 In Japan, however, since no Japanese version has been developed, we are at a disadvantage of possessing the results obtained in previous studies as a common property or executing research regarding impulsiveness.

The Barratt Impulsiveness Scale 11th version (BIS-11)10 is a short questionnaire developed for assessing impulsiveness. This instrument is the revised version of the BIS 10th version11 which has been broadly used in the world. Although there are not many studies in which the BIS-11 has been used, some studies have found that the scale was related not only to clinical findings, but also to biological findings.1,7

In the present study, we developed the Japanese version of the BIS-11 and examined its reliability and validity. We believe that the development of the Japanese version of the BIS-11 could contribute to the progress of studies on impulsiveness in Japan.

METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

Barratt Impulsiveness Scale, 11th version

The BIS-11 is a short questionnaire designed to measure impulsiveness. It contains a total of 30 items, each of which is answered on a 4-point Likert scale (rarely/never = 1, occasionally = 2, often = 3, almost always/always = 4), and a level of impulsiveness is calculated by summing up the scores for each item. All items were defined as identifying impulsiveness within the structure of related personality traits. The factor analysis revealed three components as follows: (i) ‘attentional impulsiveness’ (Iat; eight items; assessing/focusing on the task at hand, or thought insertions and racing thoughts), (ii) ‘motor impulsiveness’ (Im; 11 items; assessing acting on the spur of the moment or perseverance), and (iii) ‘non-planning impulsiveness’ (Inp; 11 items; assessing planning and thinking carefully, or enjoy challenging mental tasks). These subtraits were identified by the second-order factor analysis for the primary six factors.

Japanese version

The Japanese version of the BIS-11 was developed using the back-translation method. We originally translated the 10th version of the BIS into Japanese. Because the 10th version contains all of the questions used in the BIS-11, however, there was no problem in developing Japanese version of the BIS-11.

First, the original version was translated into Japanese by one of the authors (S.T.). After the back translation of the Japanese version was made by the other author (T. Someya), then, it was compared with the original version. If needed, corrections were made to the Japanese version in the cooperation with the native author (C.R.).

Subjects

The sample used in the present study was derived from two sources. After giving their verbal informed consent, 34 female college students, with a mean age of 19.9 years (SD 2.0), filled out the Japanese version of the BIS-11 (Sample I). The BIS-11 was re-administered to this sample 4 months later. An additional sample was volunteer workers recruited from three hospitals in the Niigata City area. After informed consent was obtained, 455 individuals were asked to complete the BIS-11. They were also asked questions regarding their demographic data. Of the subjects, 39 did not return or fill out the questionnaires. Therefore, the remaining 416 (91%) subjects were included in the present study (Sample II). They consisted of 115 men (28%) and 301 women (72%), with a mean age of 32.0 years (SD 13.1). The subjects’ duration of education was 12.8 years (SD 3.0).

Data analysis

Test–retest reliabilities of the BIS-11 and its subscales were determined in the Sample I. To see the chance-corrected agreement of the BIS scales, one-way analysis of variance intraclass coefficients (ANOVA ICC)12 was calculated. The internal consistency reliabilities of the BIS-11 and its subscales were assessed using the Sample II, by calculating Cronbach’s alpha coefficient.

The factor validity of the BIS was examined in the Sample II, using confirmatory factor analysis (CFA). Confirmatory factor analysis is a more stringent procedure for testing factor structure than exploratory factor analysis because it provides for a test significance and multiple goodness-of-fit indices. The goodness-of-fit index (GFI) and the adjusted GFI (AGFI) were used to assess the goodness-of-fit of the model. According to the recommendations by a number of studies,13–15 the following criteria were used to indicate the goodness-of-fit of a particular model: GFI > 0.85 and AGFI > 0.80.

Finally, we sought to determine whether the BIS-11 score was influenced by demographic variables (age, sex, and education).

For the analyses, the statistical software SPSS 8.0J for Windows16 (SPSS Japan Inc.) and the Amos 4.017 (SPSS Japan Inc.) were used.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

Test–retest reliability

Table 1 shows the agreement of the total BIS-11 and its subscales between test and retest. The values of ANOVA ICC ranged form 0.71 to 0.84 and all were significant.

Table 1.  Agreement of the Barratt Impulsiveness Scale between test and retest
ScalesTest mean (SD)Retest mean (SD)ANOVA ICCF
  • *

    P < 0.001.

  • BIS, Barratt Impulsiveness Scale; Iat, attentional impulsiveness; Im, motor impulsiveness; Inp, non-planning impulsiveness; ANOVA ICC, one-way analysis of variance intraclass coefficient.

Total BIS39.3 (13.6)38.4 (14.8)0.71 6.0*
Iat9.0 (4.8)8.9 (5.1)0.78 8.3*
Im12.7 (5.8)11.0 (6.5)0.74 6.6*
Inp17.6 (6.2)18.4 (6.6)0.8411.8*

Internal consistency reliability

Table 2 demonstrates Cronbach’s alpha coefficients for the total BIS-11 and its subscales. The values for the individual scales were 0.79 for the total BIS, 0.60 for Iat, 0.64 for Im, and 0.65 for Inp.

Table 2.  Internal consistency of the Barratt Impulsiveness Scale
ScaleCronbach’s alpha
  1. BIS, Barratt Impulsiveness Scale; Iat, attentional impulsiveness; Im, motor impulsiveness; Inp, non-planning

  2. impulsiveness.

Total BIS0.8
Iat0.60
Im0.64
Inp0.65

Factor validity

Since the three factors (attentional impulsiveness, motor impulsiveness, and non-planning impulsiveness) were conjectured as the main impulsiveness subtraits of the BIS-11. In the original study, the three-factor model was tested using CFA. Parameter estimates from the analysis are shown in Fig. 1. The parameter estimates among the three factors in the CFA were 0.13 between Iat and Im; 0.16 between Im and Inp; and 0.13 between Inp and Iat. The goodness-of-fit indicators suggested that the model had good fit to the data; GFI = 0.85 and AGFI = 0.82.

image

Figure 1. Factor structure of the Japanese version of the Barratt Impulsiveness Scale 11th version (BIS-11). BIS: Barratt Impulsiveness Scale; Iat, attentional impulsiveness; Im, motor impulsiveness; Inp, non-planning impulsiveness; GFI, goodness-of-fit index; AGFI, adjusted GFI.

Download figure to PowerPoint

Influence of demographic variables

A multiple regression analysis was performed to estimate the effects of age, sex and education on the BIS-11 with these factors as independent variables and a total score of the BIS-11 as a dependent variable. Table 3 shows the results of the multiple regression analysis.

Table 3.  Relationship between the Barratt Impulsiveness Scale (11th version) and demographic variables (multiple regression analysis)
Dependent variable = the BIS-11
Independent variableBBetaP
  1. BIS, Barratt Impulsiveness Scale.

Age− 0.32− 0.280.000
Gender− 0.29− 0.010.895
Education0.030.010.931

Gender and education did not contribute to the BIS-11 score while the score was significantly correlated with age.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

The results of the present study indicate that the Japanese version of the BIS-11 has sufficient reliability and validity. The stability of the instrument over time was acceptable because the chance-corrected agreement between test and retest was significantly high for all scales (ANOVA ICC = 0.71–0.84). The internal consistency of the total scale was acceptable, with Cronbach’s alpha being 0.79. The subscales had somewhat lower internal consistencies although Cronbach’s alpha for the subscales ranged between 0.6 and 0.7. However, it might result from their fewer items. The Japanese version of the BIS-11 is hence concluded to have sufficient internal consistency reliability. The result from the CFA showed that a three-model, proposed in the original study, produced acceptable fit of the data based on the goodness-of-fit indices (GFI = 0.85; AGFI = 0.82). The parameter estimates among the factors were low, indicating that the three factors are each independent. Therefore, the Japanese version of the BIS-11 is suggested to have the factor structure similar to the original one.

We also examined the relationship between the BIS-11 and several demographic variables. The results showed that gender and education variables did not correlate with the BIS-11. This suggests that the BIS-11 can consistently assess impulsiveness regardless of these variables. Age negatively correlated with the test score. In general, young persons seem to behave more impulsively than elderly persons. However, whenever the BIS-11 is used, we should consider the possibility that the score is to some extent influenced by age.

The present study was carried out using non-clinical populations. It has not yet been examined whether the Japanese version of the BIS-11 has the capability to discriminate between clinically impulsive subjects and controls. A further study is needed to confirm this.

In summary, although there are some limitations, the Japanese version of the BIS-11 is a reliable and valid measure and has possible utility for assessing impulsiveness.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES