Validity and Reliability of the Turkish Version of UPPS-P Impulsive Behavior Scale for Children (UPPS-P-C)

doi:10.21203/rs.3.rs-2216830/v1

Download PDF

Research Article

Validity and Reliability of the Turkish Version of UPPS-P Impulsive Behavior Scale for Children (UPPS-P-C)

https://doi.org/10.21203/rs.3.rs-2216830/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

In this study we aimed to examine the Turkish validity and reliability of the UPPS-P Impulsive Behavior Scale for Children (UPPS-P-C) and to investigate whether there is a relationship between the UPPS-P-C subscales and Attention Deficit Hyperactivity Disorder (ADHD).

Method

The clinical sample of the study consisted of 50 children which are aged 10–14 with ADHD who applied to the Ankara City Hospital Child Psychiatry Clinic and did not receive treatment for at least 1 month. The community sample consisted of 525 children and adolescents. They were selected from the city of Sinop and their ages ranged from 10 to 14 years, all of whom were in middle school (5th to 8th grade). UPPS-P-C and Strengths and Difficulties Questionnaire (SDQ) – parent form were used in the study.

Results

Factor analysis showed that the five-factor model better explained the data. When the relationships between the subscales of the scale were examined, all relationships were found to be statistically significant. There was a lack of premeditation and lack of perseverance that had the highest ability to distinguish children with ADHD from the UPPS-P-C subscales. In the linear regression analysis, the model we tried to explain in the SDQ total and subscale scores with the UPPS-P-C subscale was found to be significant. The Cronbach α coefficient was found to be 0.894 for the UPPS-P-C scale. When the correlation between the total and subscales of the test and retest was examined, the Intra class Correlation Coefficient (ICC) was 0.782 and 0.771–0.897, respectively.

Discussion

Our findings show that the Turkish version of the UPPS-P-C has good validity and reliability and is successful in screening for features related to impulsivity. It has been shown that the UPPS-P-C scale can be used for symptom profiling and severity assessment rather than for solely diagnostic purposes.

UPPS-P

impulsivity

validity

reliability

Impulsivity is defined as “taking quick action without forethought or conscious decision making”, “behaving without thinking sufficiently” and “the tendency to act with less thought than individuals with similar abilities and knowledge” [1]. As a personality trait, impulsivity is a feature that can affect all areas of human life and is aimed at adapting the individual [2, 3]. In addition, impulsivity has entered many psychiatric diseases and psychopathological personality models [1].

Whiteside and Lynam analyzed various previous theories of impulsivity and self-report scales and created a 4-factor model (UPPS). According to this model, the four faces of impulsivity are [4]:

Negative Urgency

Tendency to impulsive behavior in the face of stress and negative affect.

Lack of Premeditation

Acting thoughtlessly without properly considering the possible long-term consequences.

Lack of Perseverance

Difficulty staying focused on boring or difficult tasks.

Sensation Seeking

Enjoying and seeking risky, exciting and dangerous experiences.

Whiteside and Lynam defined impulsivity as an umbrella term covering these four faces (UPPS). Urgency was related to bulimia nervosa and borderline personality disorder; lack of premeditation was associated with dementia and antisocial personality disorder; lack of perseverance was related to ADHD; and sensation-seeking was associated with substance use disorders. It has been reported that UPPS may be associated with many psychopathological behaviors [4].

In 2007, Cyders et al., created a positive urgency (tendency to act thoughtlessly under extremely positive emotions) subscale, which was later included in the UPPS scale. Thus, the UPPS-P impulsivity model has been defined as a versatile and multidimensional model that includes five impulsive personality traits [5]. The 5-factor approach to impulsivity can further help us understand the formation of behaviors of an impulsive nature.

Impulsivity in children and adolescents can be seen in attention deficit hyperactivity disorder (ADHD), conduct disorder (CD) and many different mental disorders. Impulsivity can negatively affect children's and adolescents' school life, daily activities, and family and friend relationships. Since early participation in risky and maladaptive behaviors is associated with pathological life patterns, it is necessary to evaluate such behaviors very early in life, even before adolescence. For these reasons, it is very important to identify and measure impulsivity in children and adolescents [1].

Zapolski et al., developed a children's version of the UPPS-P Impulsive Behavior Scale (UPPS-P-C). The purpose of this version is to measure the UPPS-P model of impulsivity in children and adolescents [6]. While the UPPS-P-C scale is widely used in many languages, its Turkish validity and reliability have not been previously established.

The UPPS-P and UPPS-P-C impulsive behavior scales offer more dimensional assessment opportunities than the Barratt Impulsivity Scale (BIS), which is more commonly used in Turkey and has validity and reliability studied in adults. As far as we know, there is no Turkish validity and reliability study for the scales used in the assessment of impulsivity in children and adolescents in the literature. In this study, it was aimed to determine the Turkish validity and reliability of the UPPS-P-C scale in children and adolescents.

2.1. Sample

This study included clinical and community samples. The clinical sample consisted of 50 children and adolescent between the ages of 10–14, who applied to Ankara City Hospital Child Psychiatry Clinic, was diagnosed with ADHD based on DSM-5, medication-free for at least 1 month, was given approved by the parents to participate in the study.

The community sample consisted of 525 children and adolescents in the 5th-8th grades at middle schools located in the center of Sinop, was given approval by the parents to participate in the study. In order to prevent bias in student selection in the community sample, all students were tried to be reached.

Those with a history of suspected or known neurological or psychiatric disease in the clinical sample (excluding oppositional defiant disorder (ODD) and CD); in the community sample, those with a suspected or known history of neurological, psychiatric, learning or developmental disorders (including ADHD) were excluded from the study.

2.2. Method

Ethical approval was obtained from the Clinical Research Ethics Committee of the University of Health Sciences Ankara City Hospital on 19/08/2020 (Decision No: E1/2020/973). In order to conduct validity and reliability studies in Turkish children and adolescents, Gregory T. Smith, one of the developers of the UPPS-P Impulsive Behavior Scale for Children (UPPS-P-C), was contacted via e-mail and written permission was obtained. By following the standard translation and back-translation processes, the scale was given its final form and made ready for research.

Oral and written consent (with Informed Consent Form) was obtained from the students and their parents who applied to our hospital after the information about the study was given to the clinical sample group. Then, the “SDF” and the “SDQ-parent form” were filled in by the parents, and the “UPPS-P-C” was filled in by the students at the hospital. The scales filled in by the clinical sample were administered in a quiet environment. The scale was read aloud to children and adolescents before the procedure and they were told how to fill in the scale. When there were scale items they did not understand or words they did not know, they were assisted by the observer.

In the community sample, face-to-face interviews could not be held with students and parents due to the pandemic process, and information about the study was given online via the web page. Consent forms prepared by the researcher, SDF, SDQ-parent form and UPPS-P-C Impulsive Behavior Scale were filled online.

The factor validity of the scale and the relationships between the subscales of the UPPS-P-C were analyzed within the community sample (525 children and adolescents). The relationship between the UPPS-P-C scale and the SDQ, the predictive effect of the UPPS-P-C subscales on the SDQ/ADHD and internal consistency for UPPS-P-C scale were analyzed within all participants (575 children and adolescents).

To compare samples, 50 children and adolescents in the community sample were matched with ADHD patients in the clinical sample for age, gender, and parental education. In order to investigate the test-retest reliability, UPPS-P-C was re-administered to 50 different children and adolescents randomly selected from the community sample (regardless of any variable, via Excel functions). Although the time period required for re-administration was planned as 6–8 weeks, it was possible to re-administer UPPS-P-C to 50 students after approximately 13–14 weeks due to pandemic.

2.3. Data Collection Tools

Sociodemographic Data Form-SDF

It was prepared by the researchers in order to evaluate the sociodemographic characteristics of the participants.

UPPS-P Impulsive Behavior Scale for Children (UPPS-P-C)

Zapolski et al., developed a children's version of the UPPS-P Impulsive Behavior Scale (UPPS-P-C) with fewer items and one-two syllable words. UPPS-P-C Impulsive Behavior Scale consists of 40 questions. It has 5 subscales. These; the subscales are negative urgency, lack of premeditation, lack of perseverance, sensation seeking and positive urgency. Each subscale consists of 8 items. In the self-report scale, a four-point Likert system was used, in which each item was scored between 1 and 4. It has been shown by Zapolski et al., that the scale has good validity and reliability [6].

Strengths and Challenges Questionnaire (SDQ) - Parent form

It was developed by Robert Goodman in 1997 to screen for mental problems in children and young people. This questionnaire has a parent form and a school form for the ages 4–16, and an adolescent form for the ages 11–16 that the adolescent himself fills. SDQ includes 25 questions, some of which question positive and some negative behavioral characteristics. These questions are grouped under 5 sub-titles; (1) conduct problems, (2) hyperactivity/ inattention (H/I), (3) emotional problems, (4) peer relationships problems, (5) prosocial behaviors. As each title is evaluated within itself, the sum of the first four titles gives the "total difficulty score" [7]. The Turkish adaptation study was carried out by Guvenir et al., in 2008 [8].

2.4. Statistical Methods

The sociodemographic data of the participants were calculated with descriptive statistics. Mann-Whitney U test for comparison of numerical data between community and clinical samples; Pearson chi-square test, Yates chi-square test, Fisher exact test and Fisher-Freeman-Halton test were used to compare categorical data.

Factor validity of the UPPS-P-C scale was examined using Confirmatory Factor Analysis (CFA) and DWLS (Diagonally Weighted Least Square) method. In CFA, two models, five interrelated factors and these five factors explaining a quadratic factor, were tested. Fit measures for the models are given [9]. Criteria for good fit RMSEA < 0.06; CFI > 0.95; TLI > 0.95, SRMR < 0.08, and χ2/df < 3 [10, 11]. In addition, an RMSEA of less than 0.08 and greater than 0.06 was considered an acceptable fit [12]. The differences between the subscales were examined, and the Pearson correlation test was performed between the subscales.

For discriminant validity, 50 students from the community sample were matched with the clinical sample using the nearest neighbor method, taking the Propensity Scores calculated using age, gender, education level of parents and income level of the family as the distance. Then, the power of the scale score to distinguish children with ADHD was examined. For this, ROC curve analysis was applied [13]. The optimal cut-off point was determined by the Youden index. For the concurrent validity of the scale, its relationship with SDQ scores was examined. The relationships between the scores were evaluated with the Spearman correlation coefficient (r_S) [14]. The predictive power of the UPPS-P-C subscales on the SDQ was evaluated through linear regression [15]. The predictive power of UPPS-P-C on the diagnosis (ADHD) status was evaluated through logistic regression [16].

The internal consistency of the UPPS-P-C scale was examined with the Cronbach α coefficient calculated for the total and subscales of the scale. Test-retest scores were compared with the paired t-test and Wilcoxon test, depending on distribution. Test-retest reliability was evaluated with the intraclass correlation coefficient (ICC) obtained for a single measurement and absolute agreement from the mixed ANOVA model and 95% CI (Confidence Interval) [17].

Statistical significance level was accepted as p < 0.05. “lavaan” package for CFA, “semPlot” package for drawing CFA path diagram, “MatchIt” package for matching was used in RStudio (v.1.3.959) program. Other statistical analysis and calculations were made in IBM SPSS Statistics 22.0 program.

3.1. Demographic Characteristics of Participants

The mean age of the children participating in the study was 11.64 ± 1.17 years (11.60 ± 1.51 years in the community sample, 12.02 ± 1.32 years in the clinical sample). It was observed that 43.2% (n = 227) of the community sample and 72.0% (n = 36) of the clinical sample were male. The rate of students with poor (16%) and moderate (30%) academic achievement in the clinical sample was statistically higher than the other group (p < 0.05).

3.2. Validity

3.2.1. Factor Validity of the UPPS-P-C Scale

The fit indices obtained for the interrelated five-factor model (Model A) examined by CFA and the quadratic model (Model B), in which these five factors explain a quadratic factor, are given in Table 1. Accordingly, it is seen that all indices worsen in the second order model (Model B) and the five-factor model (Model A) explains the data better. The path diagram of the five-factor model is presented in Fig. 1. Factor loadings and covariance parameters between factors were found to be statistically significant (p < 0.05).

Table 1

The fit indices of the five-factor model and the quadratic model
Fit Indices	Five-Factor Model	Quadratic Model
χ²	2985.512	6250.320
df	730	735
p-value	< 0 .001	< 0.001
χ²/df	4.090	8.504
RMSEA (%95 CI)	0.073 (0.071–0.076)	0.114 (0.112–0.117)
CFI	0.970	0.926
TLI	0.968	0.921
SRMR	0.085	0.121
χ² Chi Square İndex, df Degree of Freedom, RMSEA Root Mean Square Error of Approximation, CI Confidence Interval, CFI Comparative Fit Index, TLI Turker-Lewis Index, SRMR Standardized Root Mean Square Residual.

When the relationships between the subscales of the UPPS-P-C were examined, it was observed that although all the relationships were statistically significant, the relationships between sensation seeking and other subscales, and the relationships between lack of perseverance and positive/negative urgency were negligibly weak (r_S, < 0.30). There are weak positive (r_S, 0.30–0.49) relationship between lack of premeditation and positive/negative urgency, moderately positive (r_S, 0.50–0.69) relationship between lack of premeditation and lack of perseverance, strong positive (r_S, 0.70–0.89) relationship between positive urgency and negative urgency.

3.2.2. The Distinctive Validity of the UPPS-P-C Scale

When the discriminating power of the UPPS-P-C scale scores from the clinical sample (ADHD) and the community sample was considered, it was determined that the area under the ROC curve (AUC) was greater than 0.50 for the scores except for sensation seeking (Fig. 2). Looking at the subscales, AUC was 0.841 for lack of premeditation, 0.689 for negative urgency, 0.408 for sensation seeking, 0.783 for lack of perseverance, and 0.668 for positive urgency.

3.2.3. Concurrent Validity of the UPPS-P-C Scale

Considering the relationships between the UPPS-P-C scale and the SDQ in all participants (575 children and adolescents), a moderately positive correlation was found between the UPPS-P-C lack of premeditation and lack of perseverance with the total difficulty score and the H/I subscale (r_S, 0.50–0.69, p < 0.001). The relationships between sensation seeking and SDQ scores, prosocial behaviors and UPPS-P-C scores, peer relationships problems and UPPS-P-C scores, emotional problems and lack of premeditation scores are negligibly weak (r_S, < 0.30). The relationships between the other subscales of both scales are weak (r_S, 0.30–0.49, p < 0.001).

3.2.4. Predictive Validity of the UPPS-P-C Scale

The predictive power of the UPPS-P-C subscales on the SDQ was evaluated through linear regression. The 45% change in the SDQ-Total score was explained by the variables we included in the model, the UPPS-P-C subscales (R² = .45, F(5.569) = 95.43, p < 0.001). Accordingly, lack of premeditation (β = 0.327, p < 0.001), lack of perseverance (β = 0.164, p < 0.001) and positive urgency (β = 0.224, p < .001) scores predicted the SDQ-Total scale scores.

The 49% change in SDQ-H/I subscale scores was explained by the UPPS-P-C subscales (R² = .49, F(5.569) = 111.87, p < 0.001). Accordingly, lack of premeditation (β = 0.439, p < 0.001), lack of perseverance (β = 0.206, p < 0.001) and positive urgency (β = 0.191, p < .001) scores predicted SDQ-H/I subscale scores.

The predictive power of UPPS-P-C on diagnosis (ADHD) status was evaluated through logistic regression. The regression model was statistically significant (χ2(5) = 87.85, p < 0.001). Considering the Cox&Snell R² and Nagelkerke R² values, it was seen that our model could explain a change of 31%. In addition, the Hosmer&Lemeshow test indicated that the variables were suitable for the model (χ2(8) = 7.90, p = 0.443).

The variables we added to the model covered 92.9% of all cases. When the subscales were examined, it was seen that the lack of premeditation was the subscale with a high predictive power of ADHD (p < 0.001) (Table 2).

Table 2

The predictive effect of the UPPS-P-C subscales on the ADHD
Subscales	ẞ	S.D.	Wald	p-value	Odds Ratio	%95 Confidence Interval
Lack of Premeditation	0.262	0.053	24.242	< 0.001	1.299	1.170	1.442
Negative Urgency	0.000	0.039	0.000	0.994	1.000	0.926	1.081
Sensation Seeking	-0.023	0.034	0.449	0.503	0.977	0.914	1.045
Lack of Perseverance	0.031	0.047	0.442	0.506	1.032	0.941	1.132
Positive Urgency	0.040	0.037	1.139	0.286	0.001	0.967	1.120
ẞ Beta Coefficient, SD Standard Deviation.

3.3. Reliability

In our study, internal consistency and test-retest reliability were examined to evaluate the reliability of the Turkish version of the UPPS-P-C Impulsive Behavior Scale (Table 3). The Cronbach α coefficient was found to be 0.894 for the UPPS-P-C scale. When the test-retest reliability of the scale was examined, ICC values were found to be 0.782 for the whole scale. It was seen that the 95% CIs of the ICC values were quite wide. The distribution of the scores obtained from 50 students in the test-retest application of the scale was given in Table 4. No statistically significant difference was observed between the scores (p > 0.5).

Table 3

Internal consistency coefficients and test-retest reliability for UPPS-P-C scale
Subscales	Cronbach α [n = 575]	ICC (%95 CI) [n = 50]
Lack of Premeditation	0.823	0.532 (0.300-0.705)
Negative Urgency	0.827	0.647 (0.452–0.783)
Sensation Seeking	0.771	0.642 (0.444–0.780)
Lack of Perseverance	0.822	0.624 (0.420–0.768)
Positive Urgency	0.897	0.711 (0.542–0.825)
TOTAL	0.894	0.782 (0.644–0.870)
ICC Intraclass Correlation Coefficient, CI Confidence Interval.

Table 4

The distribution of test-retest scores of the UPPS-P-C scale
	Test [n = 50]			Retest [n = 50]
Subscales	Mean ± SD	CR	Median (Q₁-Q₃)	Mean ± SD	CR	Median (Q₁-Q₃)	Test Statistic	p-value
Lack of Premeditation	13.02 ± 4.15	8–24	12 (9–16)	12.68 ± 3.57	8–23	13 (10–14)	t = 0.640	0.525
Negative Urgency	16.62 ± 5.51	8–32	17 (13–20)	16.16 ± 5.17	8–26	16 (13–19)	Z = 0.099	0.921
Sensation Seeking	22.52 ± 5.38	11–31	23 (18–28)	22.70 ± 4.93	11–31	23 (18–27)	t = 0.290	0.773
Lack of Perseverance	13.76 ± 4.55	8–29	14 (10–16)	13.70 ± 4.10	8–25	13 (10–16)	Z = 0074	0.941
Positive Urgency	15.74 ± 5.47	8–31	16 (11–19)	16.10 ± 5.95	8–32	16 (12–20)	t = 0.584	0.562
TOTAL	81.66 ± 17.15	46–129	80 (70–95)	81.34 ± 16.2	52–120	77 (70–92)	t = 0.204	0.839
SD Standard Deviation, CR Change Range (Minimum- Maximum), Q₁ 1. Quarter, Q₃ 3. Quarter.

In this study, the scale named “UPPS-P Impulsive Behavior Scale Child Version, UPPS-P-C” developed by Zapolski et al., was adapted into Turkish for the 10–14 age group, and its validity and reliability were evaluated. This study is the first to measure these characteristics in adolescence and pre-adolescent children in our country.

Firstly, two models were tested with factor analysis. Evaluating impulsivity not as a single dimension (Model B) but separately as 5 sub-dimensions (Model A) gave more appropriate results, that is, subscale scores explained impulsivity better than the total score. These models were previously examined by Geurten et al. and Pilatti et al., and similar results were shown [18, 19].

Factor loadings and covariance parameters between factors were found to be statistically significant. Accordingly, it can be said that the five factors of the Turkish version of the UPPS-P-C scale (lack of premeditation, lack of perseverance, sensation seeking, positive/negative urgency) and the items in the factor are in harmony.

The relationships between the subscales of UPPS-P-C were examined. The strong relationship between both urgency (positive/negative) subscales supports the existence of the fifth impulsivity dimension known as “positive urgency”, which was put forward by Cyders et al [5]. Similarly, in the study of Zapolski et al., a negligible weak (r_S, < 0.30) relationship was found between sensation seeking and other subscales, and between lack of perseverance and positive/negative urgency; the subscales showing the highest correlation with each other were positive and negative urgency (moderate, positive (r_S, 0.50–0.69)) [20].

The power of the UPPS-P-C scale scores to distinguish the clinical sample (ADHD) from the community sample was discussed. When the AUC values are examined, it is seen that the lack of premeditation (good level) has the highest power to distinguish children with ADHD, followed by lack of perseverance (acceptable level). In a study by Whiteside and Lynam, lack of perseverance was associated with ADHD [4]. In the meta-analysis of Berg et al., the subscales showing the highest effect size with ADHD were found to be lack of perseverance, then negative urgency and lack of premeditation. The sensation seeking subscale was not found to be significant with its negative correlation, similar to our study [21]. In our study, although the ROC analysis results of the UPPS-P-C scale were significant, it was seen that it was not sufficiently selective for diagnostic discrimination. This indicates that the UPPS-P-C scale can be used for symptom profile and severity assessment rather than for diagnostic purposes as it was designed.

In a study investigating the advanced psychometric properties of SDQ, it was shown that the H/I subgroup mean score of the group diagnosed with ADHD was statistically significantly higher than the group that did not have this clinical diagnosis [22]. Therefore, considering that the lack of premeditation and lack of perseverance subscale scores in our study were also higher in distinguishing ADHD than other subscales, the relationship between both subgroups of the scales (lack of premeditation/ lack of perseverance and H/I) seems to be an expected result.

Considering that the relations between sensation seeking and other subscales are negligibly weak, that the area under the ROC curve (AUC) is insignificant for sensation seeking, and that the correlations between sensation seeking score and SDQ scores are negligibly weak; it is understood that this subscale (sensation seeking) is useless for the ADHD group. In the meta-analysis study conducted by Berg et al., it was reported that the sensation seeking subscale showed a negative correlation close to zero with ADHD symptoms [21].

In this study, only one psychopathology associated with impulsivity (ADHD) was studied. However, each of the UPPS-P subscales is associated with different psychopathologies associated with impulsivity [4, 21, 23, 24]. It has been suggested by Whiteside and Lynam that sensation seeking may be associated with substance use disorders [4]. One study showed that urgency and sensation seeking subscales distinguished alcohol abuse patients (with and without antisocial personality traits) from controls [25]. In another study, it was shown that the subscales of lack of premeditation, lack of perseverance and sensation seeking distinguish alcohol abuse patients with borderline personality disorder, pathological gambling, and antisocial features from controls [26].

In addition, the UPPS-P scale also overlaps with major personality theories. While neuroticism in Thurstone's Big Five Theory coincides with urgency, extraversion with sensation seeking, responsibility with lack of perseverance and lack of premeditation, responsibility in NEO PI-R personality theory coincides with lack of premeditation and lack of perseverance, extraversion with sensation seeking and neuroticism with urgency [27, 28]. As a result; when we look at the literature, it has been shown in many studies that the sensation seeking subscale is valid.

In our study, as in the literature, the predictive performance of the sensation seeking subscale was shown to be insufficient for ADHD. However, this does not mean that the sensation seeking subscale is useless in other psychopathologies that have been shown to be associated (such as substance use disorders, aggression, and non-suicidal self-harm). Studies can be conducted to determine the functionality of this and other subscales and the impulsivity states they are associated with.

Since p < 0.001 was found in the linear regression analysis of our study, the model we tried to explain the SDQ total and subscale scores with the UPPS-P-C subscales is a meaningful model. According to this; in particular, it was observed that SDQ total difficulty score and H/I subscale scores could be more significantly predicted by lack of premeditation, lack of perseverance and positive urgency. In the concurrent validity analysis of our study, the finding of a positive moderate relationship between the lack of premeditation-lack of perseverance and the SDQ total-H/I subscale scores supports each other. In the logistic regression analysis, the predictive power of UPPS-P-C on the diagnosis (ADHD) status was evaluated and it was observed that the lack of premeditation was a subscale with a high predictive power for ADHD, as in the discriminant validity analysis.

The Cronbach α coefficient was found to be greater than 0.70 for the total and subscales of the UPPS-P-C scale. The internal consistency cronbach alpha value of the UPPS-P-C by Zapolski et al., was 0.84 for the lack of premeditation; 0.87 for negative urgency; 0.90 for sensation seeking; 0.81 for lack of perseverance and 0.89 for positive urgency [6]. Accordingly, the internal consistency of the UPPS-P-C scale and all its subscales is high.

For the test-retest reliability of the UPPS-P-C scale, the correlation between the total and subscales of the two tests was examined, and a moderate correlation was found for the subscale scores. It was observed that the test-retest reliability ranged from moderate to good, but the 95% confidence intervals for ICC values were quite wide. A wide confidence interval, that is, increasing the distance between the lower and upper bounds, indicates that our estimation is not sensitive. It is thought that the failure to re-apply in the planned time period in our study may have caused this situation. In the test-retest application of the scale, the total and subscale scores were similar in both applications, and no statistically significant difference was observed between them. In the Turkish adaptation study of the UPPS Impulsive Behavior Scale for Adults by Yargic et al.; similar to our study, no statistically significant difference was observed between the total and subscale mean scores of the two tests [3].

Overall, our findings show that the UPPS-P-C is a valid and reliable tool for assessing the multidimensional impulsivity structure in children. It has been shown that the UPPS-P-C scale can be used for symptom profile and severity assessment rather than for diagnostic purposes. Instead of a general measure of impulsivity, an approach that divides impulsivity into five factors seems more consistent.

Strengths of the study; it is the first scale adaptation study conducted in children in the field of impulsivity in our country, the size of the community sample, the similar age mean-distribution in the two samples, and the selection of untreated cases diagnosed with ADHD for the first time.

Comparison of two different samples in terms of geographical region and cultural environment;
The fact that the data collection process was not the same in the two groups;
The inability to differentiate between ADHD subtypes in the measurement of impulsivity;
Performing the test-retest study after 13–14 weeks, so the large 95% CIs can be affected by this situation;
The discriminant validity study was conducted only among ADHD patients and healthy subjects (not including other psychiatric diseases associated with impulsivity), so the discriminative power of the subscales from other diseases is unknown;
The level of intelligence is clinically evaluated without being measured by psychometric testing in the clinical sample, and that it is dependent on reporting in the community sample are some of the limitations of our study. Conducting new studies that do not include the limitations of our study will contribute to the use, validity and reliability of the scale.

As a result; it is thought that, thanks to the adaptation of the UPPS-P-C Impulsive Behavior Scale, which has a good validity and reliability, into Turkish, it has contributed to the examination of impulsivity and related conditions in children and adolescents in our country.

Funding

The authors did not receive support from any organization for the submitted work.

Conflict of Interest

The authors declare that there is no conflict of interest.

Availability of Data and Material

Available

Code Availability

Not applicable

Author Contribution

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Uğur Tekeoğlu. The first draft of the manuscript was written by Uğur Tekeoğlu and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Ethical Approval

This study was approved by Ankara City Hospital Ethics Committee on 19/08/2020 with the number of 2020/973.

Informed Consent

Written informed consent was obtained from the parents.

Verbal informed consent was obtained prior to the interview.

Consent for Publication

Patients signed informed consent regarding publishing their data.

Moeller FG, Barratt ES, Dougherty DM, Schmitz JM, Swann AC (2001) Psychiatric aspects of impulsivity. American Journal of Psychiatry 158:1783–1793. https://doi.org/10.1176/appi.ajp.158.11.1783
Spinella M (2004) Neurobehavioral correlates of impulsivity: evidence of prefrontal involvement. International Journal of Neuroscience 114:95–104. https://doi.org/10.1080/00207450490249347
Yargıç İ, Ersoy E, Oflaz SB (2011) UPPS dürtüsel davranış ölçeği ile psikiyatri hastalarında dürtüselliğin ölçümü. Klinik Psikofarmakoloji Bülteni-Bulletin of Clinical Psychopharmacology 21:139–146. https://doi.org/10.5455/bcp.20110706024203
Whiteside SP, Lynam DR (2001) The five factor model and impulsivity: Using a structural model of personality to understand impulsivity. Personality and Individual Differences 30:669–689. https://doi.org/10.1016/S0191-8869(00)00064-7
Cyders MA, Smith GT, Spillane NS et al (2007) Integration of impulsivity and positive mood to predict risky behavior: development and validation of a measure of positive urgency. Psychological assessment 19:107. https://doi.org/10.1037/1040-3590.19.1.107
Zapolski TCB, Stairs AM, Settles RF, Combs JL, Smith GT (2010) The measurement of dispositions to rash action in children. Assessment 17:116–125. https://doi.org/10.1177/1073191109351372
Goodman R (1997) The Strengths and Difficulties Questionnaire: a research note. Journal of Child Psychology and Psychiatry 38:581–586. https://doi.org/10.1111/j.1469-7610.1997.tb01545.x
Güvenir T, Özbek A, Baykara B et al (2008) Güçler ve güçlükler anketi'nin (GGA) Türkçe uyarlamasının psikometrik özellikleri. Turkish Journal of Child and Adolescent Mental Health 15:65–74.
Kline RB (2015) Principles and practice of structural equation modeling. Guilford publications.
Hu L, Bentler PM (1999) Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural equation modeling: a multidisciplinary journal 6:1–55. https://doi.org/10.1080/10705519909540118
Marsh HW, Hocevar D (1985) Application of confirmatory factor analysis to the study of self-concept: First-and higher order factor models and their invariance across groups. Psychological bulletin 97:562. https://doi.org/10.1037/0033-2909.97.3.562
Browne MW (1993) Alternative ways of assessing model fit. Testing structural equation models.
Mandrekar JN (2010) Receiver operating characteristic curve in diagnostic test assessment. Journal of Thoracic Oncology 5:1315–1316. https://doi.org/10.1097/JTO.0b013e3181ec173d
Stephen D, Jurs G (2003) Applied Statistics For The Behavioral Sciences.
Yan X, Su XG (2003) Linear regression analysis. Theory and Computing.
Peng CYJ, Lee KL, Ingersoll GM (2002) An introduction to logistic regression analysis and reporting. The journal of educational research 96:3–14. https://doi.org/10.1080/00220670209598786
Koo TK, Li MY (2016) A guideline of selecting and reporting intraclass correlation coefficients for reliability research. Journal of chiropractic medicine 15:155–163. https://doi.org/10.1016/j.jcm.2016.02.012
Geurten M, Catale C, Gay P, Deplus S, Billieux J (2018) Measuring impulsivity in children: adaptation and validation of a short version of the UPPS-P impulsive behaviors scale in children and investigation of its Links With ADHD. Journal of attention disorders. https://doi.org/10.1177/1087054718775831
Pilatti A, Lozano OM, Cyders MA (2015) Psychometric properties of the Spanish version of the UPPS-P Impulsive Behavior Scale: A Rasch rating scale analysis and confirmatory factor analysis. Psychological assessment 27:e10. https://doi.org/10.1037/pas0000124
Zapolski TC, Smith GT (2013) Comparison of parent versus child-report of child impulsivity traits and prediction of outcome variables. Journal of psychopathology and behavioral assessment 35:301–313. https://doi.org/10.1007/s10862-013-9349-2
Berg JM, Latzman RD, Bliwise NG, Lilienfeld SO (2015) Parsing the heterogeneity of impulsivity: A meta-analytic review of the behavioral implications of the UPPS for psychopathology. Psychological assessment 27:1129. https://doi.org/10.1037/pas0000111
Yalın Ş (2008) Türkçe güçler ve güçlükler anketi'nin ileri psikometrik özellikleri. Dokuz Eylül Üniversitesi Tıp Fakültesi.
Miller J, Flory K, Lynam D, Leukefeld C (2003) A test of the four-factor model of impulsivity-related traits. Personality and Individual Differences 34:1403–18. https://doi.org/10.1016/S0191-8869(02)00122-8
Miller JD, Lynam DR (2001) Structural models of personality and their relation to antisocial behavior: A meta-analysis. Criminology 39:765 – 98. https://doi.org/10.1111/j.1745-9125.2001.tb00940.x
Whiteside SP, Lynam DR (2003) Understanding the role of impulsivity and externalizing psychopathology in alcohol abuse: application of the UPPS impulsive behavior scale. Exp Clin Psychopharmacol 11:210–7. https://doi.org/10.1037/1064-1297.11.3.210
Whiteside SP, Lynam DR, Miller JD, Reynolds SK (2005) Validation of the UPPS impulsive behaviour scale: a four-factor model of impulsivity. European Journal of Personality 19:559–74. https://doi.org/10.1002/per.556.
Thurstone LL (1934) The vectors of the mind. Psychological Review 41:1–32. https://doi.org/10.1037/h0075959
Costa PT Jr, McCrae RR (1997) Stability and change in personality assessment: the revised NEO Personality Inventory in the year 2000. J Pers Assess 68:86–94. https://doi.org/10.1207/s15327752jpa6801_7

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Validity and Reliability of the Turkish Version of UPPS-P Impulsive Behavior Scale for Children (UPPS-P-C)

Status:

Version 1

Abstract

Background

Method

Results

Discussion

Figures

1. Introduction

2. Materials And Methods

2.1. Sample

2.2. Method

2.3. Data Collection Tools

2.4. Statistical Methods

3. Results

3.1. Demographic Characteristics of Participants

3.2. Validity

3.2.1. Factor Validity of the UPPS-P-C Scale

3.2.2. The Distinctive Validity of the UPPS-P-C Scale

3.2.3. Concurrent Validity of the UPPS-P-C Scale

3.2.4. Predictive Validity of the UPPS-P-C Scale

3.3. Reliability

4. Discussion

Declarations

References

Additional Declarations

Status:

Version 1