Revising the Epreuve de Performance Cognitive: Psychometric Properties of the Revised Nonverbal, Sequential Reasoning Test

Abstract

This study aimed to revise the Epreuve de Performance Cognitive (EPC), a nonverbal, sequential reasoning test, by incorporating a stopping requirement after five consecutive misses, and to evaluate the psychometric properties of the revised EPC. Data from 1,764 test takers were analyzed using various statistical methods. The revised EPC demonstrated high reliability, with a reliability coefficient of .94 and a Cronbach's alpha of .92. Multidimensional scaling analysis confirmed the existence of a continuum of item difficulty, and factor analysis revealed a strong relationship between the revised EPC and Scholastic Assessment Test (SAT) scores, supporting construct validity. The revised EPC also showed high correlations with other cognitive measures, indicating convergent validity. Despite some limitations, the revised EPC exhibits robust psychometric properties, making it a useful tool for assessing problem-solving ability in average and gifted adults. Future research should address study limitations and investigate the impact of timed versus liberally timed conditions on test performance.

Keywords: Epreuve de Performance Cognitive, revised EPC, nonverbal reasoning, sequential reasoning, psychometric properties, reliability, validity.

Introduction

Psychometrics is a major field within psychological research, focusing on the theory and techniques involved in psychological measurement, particularly the design, interpretation, and validation of psychological tests. The study of the psychometric properties of tests is crucial for ensuring their reliability, validity, and accuracy in assessing the intended psychological constructs. The present study aims to revise the Epreuve de Performance Cognitive (EPC), a nonverbal, sequential reasoning test, and investigate its psychometric properties.

Sequential reasoning tests are designed to assess an individual's ability to understand and predict patterns, sequences, and relationships (DeShon, Chan, & Weissbein, 1995). These tests have been widely used in different contexts, including assessing cognitive abilities, aptitude, and intelligence (Carroll, 1993). The original EPC has been employed in various research contexts, including studies on problem-solving, and giftedness (Jouve, 2005). However, the test's stopping criterion has been a topic of debate, with some researchers arguing that it may limit the test's effectiveness in distinguishing between high and low performers.

The present study aims to address this concern by adding a stopping requirement after five consecutive misses to the EPC, thereby revising the test. The rationale behind this revision is to minimize potential fatigue and frustration associated with attempting numerous difficult items without success. To assess the psychometric properties of the revised EPC, the study employs various statistical techniques, including the Spearman-Brown corrected Split-Half formula (Brown, 1910; Spearman, 1910), Cronbach's alpha (Cronbach, 1951), multidimensional scaling analysis, principal components factor analysis, and correlation analysis (Nunnally & Bernstein, 1994).

The reliability and validity of the revised EPC are of utmost importance for its potential applications in research and practice. Previous studies have utilized the EPC as a measure of cognitive abilities, such as problem-solving and time for taking the test (Jouve, 2005). Moreover, the EPC has been used in studies with gifted individuals, highlighting its potential to identify high-performing individuals (Jouve, 2005). The study's primary objective is to assess the reliability and validity of the revised EPC and compare its psychometric properties to well-established tests, such as Raven's Advanced Progressive Matrices (Raven et al., 1998), Cattell's Culture-Fair Intelligence Test-3A (Cattell & Cattell, 1973), and the Scholastic Assessment Test (SAT) (College Board, 2010).

This study seeks to address the potential limitations of the original EPC by revising the test and adding a stopping requirement after five consecutive misses. The main goal is to investigate the psychometric properties of the revised EPC, focusing on its reliability, validity, and relationship with other established cognitive measures. By providing a comprehensive analysis of the revised EPC's psychometric properties, this study aims to contribute to the literature on sequential reasoning tests and their potential applications in research and practice.

Method

Research Design

The current study utilized a quasi-experimental design to revise the Epreuve de Performance Cognitive (EPC), a nonverbal, sequential reasoning test, and to evaluate its psychometric properties (Blair & Raver, 2012). Specifically, a stopping requirement was added, where the test would be terminated after five consecutive misses, and the resulting test scores were compared to other established cognitive measures.

Participants

A total of 1,764 participants, who completed the revised EPC, were included in this study. No exclusion criteria were set.

Materials

The revised EPC was employed as the primary measure for this study. The original EPC is a nonverbal, sequential reasoning test that assesses problem-solving ability (Jouve, 2005). Modifications to the original EPC included the addition of a stopping requirement after five consecutive misses. The revised EPC was then compared to other well-established cognitive measures, such as Raven's Advanced Progressive Matrices (APM; Raven, 1998), Cattell's Culture-Fair Intelligence Test-3A (CFIT; Cattell & Cattell, 1973), and the Wechsler Adult Intelligence Scale (WAIS; Wechsler, 1997), to establish convergent validity.

Procedures

Upon obtaining informed consent, participants were administered the revised EPC individually. The revised EPC is computerized and consisted of 35 items, with participants instructed to complete as many items as possible under liberally timed conditions. To ensure data quality, the stopping requirement was implemented, terminating the test after five consecutive misses. For the convergent validity studies, participants were then asked to complete the APM, CFIT, or WAIS. When possible, participants were asked to report their previous scores, especially on college admission tests, such as the SAT.

Statistical Analyses

The data were analyzed using various statistical techniques, such as the Spearman-Brown corrected Split-Half formula, Cronbach's alpha, multidimensional scaling analysis, principal components factor analysis, and correlation analysis. The Spearman-Brown formula and Cronbach's alpha were used to assess the reliability of the revised EPC scores, while multidimensional scaling analysis was employed to examine the test's structure with ALSCAL (Young et al., 1978). Principal components factor analysis was conducted to establish construct validity, and correlation analysis was used to determine the convergent validity of the revised EPC with other cognitive measures. Apart from MDS, all the analyses were carried out with Excel.

Results

Statistical Analyses

The goal of this study was to revise the Epreuve de Performance Cognitive (EPC), a nonverbal, sequential reasoning test, by adding a stopping requirement after five consecutive misses, and to examine the psychometric properties of the revised EPC. The data collected from 1,764 test takers were analyzed using various statistical tests, including the Spearman-Brown corrected Split-Half formula, Cronbach's alpha, multidimensional scaling analysis, principal components factor analysis, and correlation analysis.

Reliability of the Revised EPC

The reliability of the scores yielded by the revised EPC was assessed using the Spearman-Brown corrected Split-Half formula and Cronbach's alpha. The entire sample of 1,764 test takers yielded a reliability coefficient of .94, calculated using the Spearman-Brown formula, indicating a high level of internal consistency. Additionally, Cronbach's alpha was found to be .92, further supporting the reliability of the revised EPC.

Multidimensional Scaling Analysis

A multidimensional scaling analysis was conducted to confirm the existence of a continuum in items from the easiest to the hardest. The two-dimensional solution appeared in a typical horseshoe shape, as shown in Figure 1, with a Stress value of .14 and an RSQ of .92. These results suggest that the revised EPC has a coherent structure in terms of item difficulty.

Figure 1. Two-dimensional scaling for the Items of the Revised EPC.

Note. N = 1,764. Root Squared Mean (RSQ) = .92. Kruskal's Stress = .14.

Factor Analysis

A principal components factor analysis was performed using the data of 95 participants who reported recentered Scholastic Assessment Test (SAT) scores. The first unrotated factor loading for the revised EPC was .83. The Math reasoning scale of the SAT loaded at .82, and the Verbal reasoning part at .75. This indicates that the EPC shares considerable variance with the SAT, supporting its construct validity.

Correlations with Other Measures

The revised EPC raw scores were found to have high correlations with other cognitive measures. A correlation of .82 was observed between the EPC raw scores and the Raven's Advanced Progressive Matrices (APM) in a sample of 134 subjects, while a correlation of .81 was found between the EPC raw scores and the Cattell's Culture-Fair Intelligence Test-3A (CFIT) in a sample of 156 observations. Additionally, a correlation of .85 was found between the EPC raw scores and the Full Scale IQ (FSIQ) on the Wechsler Adult Intelligence Scale (WAIS) in a highly selective sample of 23 adults with an average FSIQ of 131.70 (SD=24.35). These results demonstrate the convergent validity of the revised EPC.

Limitations

Despite the promising results, some limitations should be considered. First, the sample size of certain sub-analyses (e.g., the correlation with FSIQ on the WAIS) was relatively small, which may limit the generalizability of the findings. Second, the study did not explore potential differences between timed and liberally timed conditions, which could provide further insight into the performance of the revised EPC.

The revised EPC, with the addition of a stopping requirement after five consecutive misses, demonstrated strong psychometric properties, including high reliability and convergent validity. The multidimensional scaling analysis confirmed the existence of a continuum in items from the easiest to the hardest, and the factor analysis demonstrated the construct validity of the revised EPC in relation to the SAT. These results support the utility of the revised EPC for assessing problem-solving ability in individuals of average ability level and gifted adults. Further research should address the limitations of the current study and explore the potential impact of timed versus liberally timed conditions on the revised EPC performance.

Discussion

Interpretation of Study Results and Relation to Previous Research

The main objective of this study was to revise the Epreuve de Performance Cognitive (EPC; Jouve, 2005) by implementing a stopping requirement after five consecutive misses and to evaluate its psychometric properties. The results indicate that the revised EPC possesses high reliability, as demonstrated by a Spearman-Brown corrected Split-Half formula coefficient of .94 and a Cronbach's alpha of .92. These findings align with previous research emphasizing the importance of test reliability in psychological assessments (Nunnally, 1978).

The multidimensional scaling analysis revealed a coherent structure in terms of item difficulty, confirming the existence of a continuum from the easiest to the hardest items. This result is consistent with prior studies that have employed multidimensional scaling analysis to identify the underlying structure of cognitive test items (Thiébaut, 2000). Furthermore, the factor analysis indicated that the revised EPC shares substantial variance with the SAT (College Board, 2010), thus supporting its construct validity. These findings are in line with previous research establishing the validity of cognitive tests in measuring problem-solving abilities (Carroll, 1993).

Implications for Theory, Practice, and Future Research

The strong psychometric properties of the revised EPC, including its high reliability and convergent validity, have significant implications for both theory and practice. The revised EPC can serve as a useful tool for assessing problem-solving ability in individuals of average ability level and gifted adults, potentially informing educational and occupational decision-making processes (Lubinski & Benbow, 2006). Moreover, the positive relationship between the revised EPC and established cognitive measures, such as the SAT, Raven's APM, CFIT, and WAIS, further substantiates the relevance of nonverbal, sequential reasoning tests in cognitive assessment (Sternberg, 2003).

Given the current findings, future research could explore the impact of time constraints on EPC performance, as the present study did not investigate potential differences between timed and liberally timed conditions. Additionally, researchers could examine the applicability of the revised EPC in diverse populations and settings, such as in clinical or cross-cultural contexts (Van de Vijver & Tanzer, 2004).

Limitations and Alternative Explanations

Despite the promising results, this study has some limitations that may affect the generalizability of the findings. First, the sample size for certain sub-analyses (e.g., the correlation with FSIQ on the WAIS) was relatively small, potentially limiting the robustness of these results (Cohen, 1988). Second, the study did not investigate the potential impact of timed versus liberally timed conditions on the revised EPC performance, which could provide valuable insights into the test's utility in various contexts (Ackerman & Kanfer, 2009).

Future Directions

The revised EPC, with the addition of a stopping requirement after five consecutive misses, demonstrated strong psychometric properties, including high reliability and convergent validity. The findings support the utility of the revised EPC in assessing problem-solving ability in individuals of average ability level and gifted adults. Future research should address the limitations of the current study, explore the potential impact of timed versus liberally timed conditions on the revised EPC performance, and investigate its applicability in diverse populations and settings (Sackett & Wilk, 1994).

Conclusion

This study successfully revised the Epreuve de Performance Cognitive (EPC) by adding a stopping requirement after five consecutive misses and demonstrated strong psychometric properties for the revised test. The reliability and convergent validity of the revised EPC were found to be high, and the multidimensional scaling analysis supported its coherent structure regarding item difficulty. Additionally, the factor analysis showed a strong relationship between the revised EPC and SAT scores, further establishing its construct validity.

These findings have important implications for the broader field of cognitive assessment, as the revised EPC offers a reliable and valid measure of problem-solving abilities for both average ability-level individuals and gifted adults. However, this study has some limitations, such as relatively small sample sizes in certain sub-analyses and the lack of investigation into potential differences between timed and liberally timed conditions.

Future research should address these limitations and explore the impact of timing conditions on the revised EPC performance. Overall, the revised EPC presents a valuable tool for cognitive assessment, and its continued refinement and investigation will contribute to the advancement of the field.

References

Ackerman, P. L., & Kanfer, R. (2009). Test length and cognitive fatigue: an empirical examination of effects on performance and test-taker reactions. Journal of experimental psychology. Applied, 15(2), 163–181. https://doi.org/10.1037/a0015719

Brown, W. (1910). Some Experimental Results in the Correlation of Mental Abilities. British Journal of Psychology, 3, 296-322. https://doi.org/10.1111/j.2044-8295.1910.tb00207.x

Carroll, J. B. (1993). Human cognitive abilities: A survey of factor-analytic studies. New York: Cambridge University Press. https://doi.org/10.1017/CBO9780511571312

Cattell, R. B., & Cattell, A. K. S. (1973). Technical supplement for the culture fair intelligence tests: Scales 2 and 3. Champaign, IL: Institute for Personality and Ability Testing.

Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates.

College Board. (2010). The SAT® test: Overview. Retrieved from https://collegereadiness.collegeboard.org/sat

Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests. Psychometrika, 16(3), 297-334. https://doi.org/10.1007/BF02310555

DeShon, R. P., Chan, D., & Weissbein, D. A. (1995). Verbal overshadowing effects on Raven's Advanced Progressive Matrices: Evidence for multidimensional performance determinants. Intelligence, 21(2), 135–155. https://doi.org/10.1016/0160-2896(95)90023-3

Jouve, X. (2005). Epreuve de Performance Cognitive (EPC). Paris, FR: Editions du Centre de Psychologie Appliquée.

Lubinski, D., & Benbow, C. P. (2006). Study of Mathematically Precocious Youth After 35 Years: Uncovering Antecedents for the Development of Math-Science Expertise. Perspectives on Psychological Science, 1(4), 316–345. https://doi.org/10.1111/j.1745-6916.2006.00019.x

Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory (3rd ed.). New York: McGraw-Hill. https://doi.org/10.1177/014662169501900308

Raven, J., Raven, J. C., & Court, J. H. (1998). Raven Manual: Section 4, Advanced Progressive Matrices, 1998 Edition. Oxford, UK: Oxford Psychologists Press.

Sackett, P. R., & Wilk, S. L. (1994). Within-group norming and other forms of score adjustment in preemployment testing. American Psychologist, 49(11), 929–954. https://doi.org/10.1037/0003-066X.49.11.929

Spearman, C. (1910). Correlation calculated from faulty data. British Journal of Psychology, 3(3), 271-295. https://doi.org/10.1111/j.2044-8295.1910.tb00206.x

Sternberg, R. J. (2003). A Broad View of Intelligence: The Theory of Successful Intelligence. Consulting Psychology Journal: Practice and Research, 55(3), 139–154. https://doi.org/10.1037/1061-4087.55.3.139

Thiébaut, E. (2000). Les Bonnardel: les tests de raisonnement [B53 et BLS4]. Paris, FR: Editions et Applications Psychologiques.

van de Vijver, F. J., & Tanzer, N. K. (2004). Bias and equivalence in cross-cultural assessment: An overview. European Review of Applied Psychology, 54(2), 119-135. http://dx.doi.org/10.1016/j.erap.2003.12.004

Wechsler, D. (1997). Wechsler Adult Intelligence Scale (3rd ed.). San Antonio, TX: Psychological Corporation.

Young, F. W., Takane, Y., & Lewyckyj, R. (1978). ALSCAL: A nonmetric multidimensional scaling program with several individual-differences options. Behavior Research Methods & Instrumentation, 10(3), 451–453. https://doi.org/10.3758/BF03205177