The reliability and validity of the six traits comprising the High Potential Trait Indicator (HPTi) were evaluated using Rasch analysis. Focus was designated to the unidimensionality and local independence of each subscale; fit to the Rasch model; person reliability and separation; and differential item functioning (DIF). Secondary data, obtained from intellectual property rights holder Thomas International, were used for analysis with a sample of 1257 South African respondents. One of the six traits,
This study highlighted the shortcomings of the current HPTi in the South African context through Rasch analysis. The findings illustrate the difficult nature of creating ideal personality instruments in the South African context, thus contributing to the body of knowledge of personality assessments in South Africa.
The use of psychometric testing in decision-making is commonplace in various sectors. Sectors include education, human resources, coaching, forensics, counselling, medical and clinical applications and economic and financial sectors (Arráiz et al.,
The HPTi is a self-reporting six-trait personality-based questionnaire with a seven-point Likert-type scale. It was developed in the United Kingdom to identify high performers (MacRae & Furnham,
In its initial development, the six HPTi scales achieved sufficient internal consistency reliability, with alpha coefficients above 0.70 (MacRae & Furnham,
According to MacRae and Furnham (
Finally, MacRae and Furnham (
Rasch measurement theory (RMT; Rasch,
(a) produce linear measures, (b) overcome missing data, (c) give estimates of precision, (d) have devices for detecting misfitting items/persons, and (e) the parameters of the object being measured and of the measurement instrument must be separable (Wright and Mok
To satisfy the requirements of measurement from a Rasch perspective and psychometric evaluation, the analyses, then, are to evaluate the person reliability, fit to the Rasch model and differential item functioning (DIF) (Bond et al.,
Rasch analysis provides two fit statistics for persons and items: infit and outfit statistics. Winsteps (Linacre,
A reasonable fit statistic range for rating scales, such as Likert-type scales, contains MNSQ values between 0.6 and 1.4 (Wright & Linacre,
According to Linacre (
The reliability of an instrument is its degree of consistency at measuring what it purports to measure (Roodt & De Kock,
In Rasch measurement, internal consistency reliability is reported as two metrics: person reliability and person separation (Bond et al.,
Unidimensionality and local independence are two interrelated conditions required for Rasch measurement (Fan & Bond,
Local independence is the condition in which an individual’s responses to an item is not affected by their response to any other items (Fan & Bond,
Differential item functioning is an evaluation of how congruently the items of a measure define a construct between certain groups (Boone et al.,
Demographic statistics of the sample.
Variable | % | |
---|---|---|
Total | 1257 | - |
Male | 573 | 45.6 |
Female | 684 | 54.4 |
Black | 380 | 30.2 |
Mixed race | 184 | 14.6 |
Asian and Indian | 106 | 8.4 |
White | 577 | 45.9 |
Other | 10 | 0.8 |
English | 558 | 44.4 |
Afrikaans | 370 | 29.4 |
Sesotho | 48 | 3.8 |
isiXhosa | 76 | 6.0 |
isiZulu | 59 | 4.7 |
Setswana | 49 | 3.9 |
Sepedi | 47 | 3.7 |
Xitsonga | 16 | 1.3 |
isiNdebele | 2 | 0.2 |
siSwati | 9 | 0.7 |
Tshivenda | 20 | 1.6 |
French | 3 | 0.2 |
Eastern Cape | 80 | 6.4 |
Free State | 123 | 9.8 |
Gauteng | 586 | 46.6 |
KwaZulu-Natal | 112 | 8.9 |
Limpopo | 13 | 1.0 |
Mpumalanga | 15 | 1.2 |
North West | 10 | 0.8 |
Northern Cape | 5 | 0.4 |
Western Cape | 313 | 24.9 |
Regarding ethnic background, slightly less than half reported as white (
When indicating their home language, 558 (44.4%) reported English and 370 (29.4%) indicated Afrikaans as their home language. IsiXhosa was the next most frequently reported home language with 76 (6.0%) respondents, followed by isiZulu (
Nearly half of the respondents reported Gauteng (
Descriptive statistics of the years of birth of respondents.
Statistics | Year |
---|---|
Median | 1982 |
Mode | 1985 |
Oldest | 1945 |
Youngest | 1999 |
Range | 54 |
Secondary data were obtained from Thomas International Ltd – the intellectual property right holder of the HPTi. The dataset includes the raw data, with scores from strongly disagree (1) to strongly agree (7) of 1257 individuals. The participants completed the HPTi for various purposes, including third-party recruitment and research conducted by Thomas International Ltd. Only data of respondents who completed the HPTi through the South African division of the organisation and had indicated their voluntary participation in further research were obtained. Negatively phrased items were reverse scored.
The primary data analyses were conducted in Winsteps (Linacre,
Descriptive statistics were examined using Microsoft Excel to outline the trends in the demographics of the sample (
Descriptive statistics of person measure scores and reliability indices of Each High Potential Trait Indicator trait.
Trait | Mean | SD | Median | Minimum | Maximum | Skewness | Kurtosis | Alpha | P. Reliability | P. Separation |
---|---|---|---|---|---|---|---|---|---|---|
CN | 1.30 | 0.77 | 1.16 | 0.00 | 4.89 | 2.06 | 6.19 | 0.68 | 0.65 | 1.37 |
AJ | 0.82 | 0.57 | 0.76 | 0.63 | 4.81 | 1.56 | 6.58 | 0.75 | 0.69 | 1.49 |
CU | 1.17 | 0.72 | 1.06 | −0.33 | 5.47 | 1.65 | 5.76 | 0.75 | 0.73 | 1.62 |
RA | 0.61 | 0.48 | 0.58 | −0.46 | 3.82 | 1.43 | 5.12 | 0.69 | 0.64 | 1.33 |
AA | −0.07 | 0.37 | −0.10 | −1.28 | 1.30 | 0.19 | 0.66 | 0.68 | 0.65 | 1.35 |
CM | 0.06 | 0.40 | 0.07 | −1.61 | 1.89 | 0.24 | 1.70 | 0.72 | 0.69 | 1.48 |
SD, standard deviation; AA, Ambiguity Acceptance; AJ, Adjustment; CM, Competitiveness; CN, Conscientiousness; CU, Curiosity; RA, Risk Approach; Alpha, Cronbach’s alpha; P. Reliability, Person Reliability; P. Separation, Person Separation.
The person reliability indices of each trait were evaluated to which a person reliability of 0.70 and separation of 1.50 are regarded as sufficiently reliable.
Misfitting items in the item fit analysis for each HPTi trait were detected and labelled as either underfitting or overfitting the model based on the infit statistics: 0.60 ≥ mean squared (MNSQ) ≥ 1.40 and z-standardised (ZSTD) ≥ |2|.
The unidimensionality of each HPTi scale was evaluated through PCAR in Winsteps (Linacre,
Item statistics and fit status.
Item | Measure | 1c Load. | IN.MNSQ | IN.ZSTD | OUT.MNSQ | OUT.ZSTD | Fit status |
---|---|---|---|---|---|---|---|
CN01 | −0.71 | 0.03 | 1.42 | 5.32 | 1.41 | 5.18 | Underfit |
CN02 | −0.32 | 0.18 | 0.87 | −2.01 | 0.87 | −2.07 | Fit |
CN03 | 0.42 | −0.24 | 1.01 | 0.20 | 1.15 | 2.74 | Fit |
CN04 | −1.04 | 0.54 | 0.92 | −1.14 | 0.65 | −5.27 | Fit |
CN05 | −0.07 | 0.12 | 0.96 | −0.67 | 1.06 | 0.97 | Fit |
CN06 | 0.54 | −0.14 | 1.02 | 0.57 | 1.20 | 3.91 | Fit |
CN07 | 0.08 | 0.62 | 0.84 | −3.02 | 0.91 | −1.55 | Fit |
CN08 | −0.35 | 0.75 | 0.88 | −1.90 | 0.74 | −4.32 | Fit |
CN09 | 0.10 | −0.04 | 1.13 | 2.36 | 1.26 | 4.11 | Fit |
CN10 | 0.06 | −0.16 | 2.02 | 9.90 | 2.14 | 9.90 | Underfit |
CN11 | −0.19 | 0.24 | 0.73 | −4.75 | 0.74 | −4.49 | Fit |
CN12 | 0.90 | −0.33 | 1.01 | 0.20 | 1.15 | 3.41 | Fit |
CN13 | 0.57 | −0.37 | 1.14 | 3.28 | 1.19 | 3.82 | Fit |
AJ01 | 0.26 | −0.45 | 0.70 | −8.90 | 0.72 | −7.28 | Fit |
AJ02 | −0.73 | −0.47 | 1.29 | 4.33 | 1.07 | 1.14 | Fit |
AJ03 | −0.75 | −0.41 | 1.24 | 3.51 | 1.14 | 2.12 | Fit |
AJ04 | 0.59 | 0.46 | 1.08 | 2.27 | 1.14 | 3.61 | Fit |
AJ05 | 0.04 | −0.31 | 0.89 | −2.67 | 1.00 | 0.09 | Fit |
AJ06 | −0.32 | −0.31 | 1.42 | 7.19 | 1.31 | 5.09 | Underfit |
AJ07 | 0.17 | 0.45 | 1.20 | 4.87 | 1.26 | 5.36 | Fit |
AJ08 | −0.42 | −0.06 | 1.00 | 0.00 | 0.94 | −1.06 | Fit |
AJ09 | 0.52 | −0.43 | 0.89 | −3.25 | 1.01 | 0.21 | Fit |
AJ10 | 0.39 | 0.05 | 1.19 | 4.82 | 1.30 | 6.65 | Fit |
AJ11 | −0.07 | 0.16 | 0.96 | −0.96 | 1.08 | 1.53 | Fit |
AJ12 | 0.41 | 0.56 | 1.19 | 4.96 | 1.35 | 7.84 | Fit |
AJ13 | −0.08 | 0.10 | 0.90 | −2.40 | 0.87 | −2.72 | Fit |
CU01 | −0.05 | 0.21 | 0.81 | −4.35 | 0.84 | −3.59 | Fit |
CU02 | −0.52 | 0.45 | 0.73 | −5.74 | 0.69 | −6.53 | Fit |
CU03 | 0.81 | −0.52 | 1.10 | 2.62 | 1.21 | 5.06 | Fit |
CU04 | −0.93 | 0.41 | 0.92 | −1.54 | 0.86 | −2.47 | Fit |
CU05 | −0.86 | 0.64 | 0.92 | −1.47 | 0.85 | −2.71 | Fit |
CU06 | 0.46 | 0.09 | 1.24 | 5.62 | 1.25 | 5.53 | Fit |
CU07 | −0.73 | 0.58 | 0.80 | −3.97 | 0.74 | −5.08 | Fit |
CU08 | 0.14 | 0.09 | 1.18 | 3.93 | 1.33 | 6.57 | Fit |
CU09 | −0.20 | 0.09 | 1.03 | 0.54 | 1.07 | 1.40 | Fit |
CU10 | 0.29 | −0.12 | 0.97 | −0.66 | 1.13 | 2.91 | Fit |
CU11 | 1.06 | −0.44 | 1.25 | 6.47 | 1.30 | 7.32 | Fit |
CU12 | −0.39 | 0.57 | 0.80 | −4.20 | 0.81 | −4.01 | Fit |
CU13 | 0.93 | −0.50 | 1.32 | 7.95 | 1.45 | 9.90 | Fit |
RA01 | 0.31 | −0.26 | 0.93 | −2.07 | 1.03 | 0.69 | Fit |
RA02 | 0.48 | −0.60 | 1.31 | 8.20 | 1.39 | 9.35 | Fit |
RA03 | −0.13 | 0.62 | 0.81 | −4.60 | 0.80 | −4.45 | Fit |
RA04 | 0.59 | 0.38 | 0.86 | −4.48 | 0.95 | −1.37 | Fit |
RA05 | 0.37 | 0.19 | 0.82 | −5.29 | 0.87 | −3.42 | Fit |
RA06 | −0.04 | −0.18 | 1.19 | 4.15 | 1.31 | 6.21 | Fit |
RA07 | −0.42 | 0.45 | 0.86 | −2.89 | 0.88 | −2.35 | Fit |
RA08 | −0.18 | −0.22 | 1.28 | 5.63 | 1.36 | 6.77 | Fit |
RA09 | 0.22 | −0.19 | 1.16 | 4.07 | 1.32 | 7.09 | Fit |
RA10 | −0.70 | 0.48 | 0.87 | −2.31 | 0.79 | −4.05 | Fit |
RA11 | −0.22 | −0.14 | 0.86 | −3.22 | 0.89 | −2.22 | Fit |
RA12 | −0.31 | 0.34 | 0.88 | −2.57 | 1.03 | 0.54 | Fit |
RA13 | 0.03 | −0.15 | 1.44 | 9.37 | 1.62 | 9.90 | Underfit |
AA01 | −0.60 | −0.10 | 1.48 | 9.90 | 1.44 | 9.15 | Underfit |
AA02 | 0.99 | 0.47 | 1.06 | 1.18 | 1.03 | 0.52 | Fit |
AA03 | −0.42 | −0.11 | 1.05 | 1.36 | 1.04 | 1.02 | Fit |
AA04 | 0.06 | 0.12 | 0.82 | −5.80 | 0.82 | −5.55 | Fit |
AA05 | 0.17 | 0.07 | 1.25 | 6.86 | 1.28 | 7.35 | Fit |
AA06 | −0.30 | −0.26 | 0.96 | −1.23 | 0.94 | −1.74 | Fit |
AA07 | 0.09 | 0.31 | 0.75 | −8.32 | 0.76 | −7.64 | Fit |
AA08 | −0.77 | −0.63 | 0.96 | −0.97 | 0.98 | −0.33 | Fit |
AA09 | 0.32 | −0.07 | 0.93 | −1.96 | 0.95 | −1.27 | Fit |
AA10 | 0.39 | 0.54 | 0.80 | −5.83 | 0.79 | −5.74 | Fit |
AA11 | 0.08 | −0.07 | 1.08 | 2.45 | 1.14 | 3.74 | Fit |
AA12 | 0.56 | 0.68 | 0.69 | −8.81 | 0.66 | −8.98 | Fit |
AA13 | −0.58 | −0.54 | 1.39 | 8.88 | 1.54 | 9.90 | Underfit |
CM01 | −0.14 | −0.24 | 0.84 | −4.65 | 0.83 | −4.91 | Fit |
CM02 | −2.04 | −0.30 | 1.34 | 4.57 | 1.25 | 3.49 | Fit |
CM03 | 0.11 | −0.21 | 0.78 | −6.94 | 0.83 | −5.21 | Fit |
CM04 | 0.26 | −0.39 | 0.88 | −3.86 | 0.90 | −2.98 | Fit |
CM05 | 0.01 | −0.52 | 1.01 | 0.17 | 1.00 | 0.15 | Fit |
CM06 | 0.08 | −0.18 | 0.99 | −0.44 | 1.05 | 1.47 | Fit |
CM07 | 0.21 | 0.50 | 1.07 | 2.06 | 1.14 | 3.85 | Fit |
CM08 | 0.16 | 0.35 | 1.17 | 4.92 | 1.20 | 5.31 | Fit |
CM09 | 0.64 | 0.60 | 0.94 | −1.64 | 0.98 | −0.41 | Fit |
CM10 | 0.42 | 0.15 | 1.14 | 3.90 | 1.15 | 3.92 | Fit |
CM11 | 0.26 | 0.37 | 0.98 | −0.69 | 0.99 | −0.17 | Fit |
CM12 | −0.38 | −0.21 | 0.90 | −2.55 | 0.87 | −3.38 | Fit |
CM13 | 0.42 | −0.12 | 1.32 | 8.54 | 1.34 | 8.45 | Fit |
AA, Ambiguity Acceptance; AJ, Adjustment; CM, Competitiveness; CN, Conscientiousness; CU, Curiosity; RA, Risk Approach; 1c Load., first contrast loading.
Evidence for local independence of each HPTi scale items was based on correlations between item residuals. From these estimates, positive correlations between items of 0.7 or higher are considered to be in violation of local independence (Fan & Bond,
The DIF was analysed across the items of each HPTi trait and between the relevant subgroups by means of a significant difference in the Rasch–Welch statistic and a sufficiently large DIF contrast. A
The subgroups for the DIF analysis are gender, ethnicity and language. However, because of the differences in sample sizes between the European languages (English and Afrikaans) and African languages (Sesotho, isiXhosa, isiZulu, Setswana, Sepedi, Xitsonga, isiNdebele and siSwati), the African languages were collapsed to form the group ‘African Languages’. The resultant language groups are thus African languages (
The secondary data obtained from Thomas International Ltd (thomas.co) contained the anonymised responses of individuals who completed the HPTi and indicated their voluntary participation in further research. Respondents were presented with the opportunity to indicate their voluntary participation in further research after completing the HPTi. Ethical approval was obtained from the Psychology Research and Ethics Committee at the University of Pretoria (reference number: HUM037/0720).
The reliability indices ranged from adequate to inadequate (see
Outfit and infit statistics were evaluated for the items of the HPTi traits, with precedence given to infit. The infit and outfit statistics of the items can be viewed in
The unidimensionality of each scale was examined through PCAR. The item loadings of the first contrast can be seen in
The largest standardised residual correlations were analysed to evaluate the local independence of the items of the scale. No item pairs were found to be above the correlation of 0.70, indicating that none of the items of the scales are in violation of local independence.
The analysis of DIF on gender revealed no items in concern across all HPTi scales. While levels of significance were detected, the significant items are not described because the DIF effect sizes of these items were negligible, indicating no practical significance.
The DIF between gender groups.
Item | Mean (S.E.) |
DIF contrast | Effect size | ||
---|---|---|---|---|---|
Female | Male | ||||
CN01 | −0.81 (0.06) | −0.60 (0.06) | 0.02 | −0.21 | Negligible |
CN02 | −0.40 (0.05) | −0.22 (0.05) | 0.01 | −0.19 | Negligible |
CN05 | 0.00 (0.04) | −0.16 (0.05) | 0.01 | 0.15 | Negligible |
CN08 | −0.44 (0.05) | −0.25 (0.05) | 0.01 | −0.19 | Negligible |
CN10 | 0.12 (.04) | −0.01 (0.04) | 0.02 | 0.13 | Negligible |
CN11 | −0.11 (0.04) | −0.28 (0.05) | 0.01 | 0.18 | Negligible |
AJ03 | −0.05 (0.03) | 0.15 (0.03) | 0.00 | −0.21 | Negligible |
AJ05 | −0.26 (0.03) | −0.40 (0.04) | 0.01 | 0.15 | Negligible |
AJ06 | −0.49 (0.04) | −0.33 (0.04) | 0.00 | −0.16 | Negligible |
AJ08 | 0.58 (0.03) | 0.45 (0.03) | 0.00 | 0.13 | Negligible |
AJ09 | 0.43 (0.03) | 0.33 (0.03) | 0.02 | 0.10 | Negligible |
CU01 | 0.01 (0.04) | −0.12 (0.04) | 0.02 | 0.14 | Negligible |
CU02 | −0.45 (0.04) | −0.61 (0.05) | 0.02 | 0.16 | Negligible |
RA12 | −0.35 (0.03) | −0.25 (0.04) | 0.04 | −0.11 | Negligible |
RA13 | 0.07 (0.03) | −0.02 (0.03) | 0.05 | 0.09 | Negligible |
AA01 | −0.64 (0.03) | −0.54 (0.03) | 0.02 | −0.09 | Negligible |
AA07 | 0.14 (0.03) | 0.04 (0.03) | 0.01 | 0.10 | Negligible |
CM03 | 0.06 (0.02) | 0.17 (0.03) | 0.00 | −0.11 | Negligible |
CM04 | 0.31 (0.03) | 0.21 (0.03) | 0.01 | 0.10 | Negligible |
CM05 | 0.10 (0.02) | −0.11 (0.03) | 0.00 | 0.21 | Negligible |
CM06 | 0.00 (0.03) | 0.17 (0.03) | 0.00 | −0.16 | Negligible |
CM07 | 0.16 (0.02) | 0.27 (0.03) | 0.00 | −0.11 | Negligible |
CM08 | 0.22 (0.02) | 0.08 (0.03) | 0.00 | 0.14 | Negligible |
DIF, differential item functioning; S.E., standard error; AA, Ambiguity Acceptance; AJ, Adjustment; CM, Competitiveness; CN, Conscientiousness; CU, Curiosity; RA, Risk Approach.
The presence of DIF was evaluated between ethnicities.
Differential item functioning between ethnicity groups.
Item | Group 1 | Mean (S.E.) | Group 2 | Mean (S.E.) | DIF contrast | Effect size | |
---|---|---|---|---|---|---|---|
CN04 | A/I | −1.41 (0.22) | W | −0.81 (0.07) | 0.00 | −0.59 | Slight to moderate |
CN04 | B | −1.45 (0.12) | C | −0.98 (0.13) | 0.03 | −0.47 | Slight to moderate |
CN04 | B | −1.45 (0.12) | W | −0.81 (0.07) | 0.00 | −0.63 | Moderate to large |
CN08 | A/I | −0.62 (0.15) | W | −0.19 (0.05) | 0.01 | −0.43 | Slight to moderate |
CU02 | B | −0.87 (0.08) | W | −0.35 (0.04) | 0.00 | −0.52 | Slight to moderate |
CU04 | A/I | −1.23 (0.16) | C | −0.73 (0.09) | 0.01 | −0.51 | Slight to moderate |
CU04 | A/I | −1.23 (0.16) | W | −0.77 (0.05) | 0.01 | −0.47 | Slight to moderate |
CU04 | B | −1.38 (0.10) | C | −0.73 (0.09) | 0.00 | −0.65 | Moderate to large |
CU04 | B | −1.38 (0.10) | W | −0.77 (0.05) | 0.00 | −0.61 | Slight to moderate |
CU06 | A/I | 0.67 (0.08) | B | 0.22 (0.05) | 0.00 | 0.45 | Slight to moderate |
CU13 | A/I | 0.76 (0.08) | B | 1.29 (0.04) | 0.00 | −0.53 | Slight to moderate |
CU13 | B | 1.29 (0.04) | C | 0.85 (0.06) | 0.00 | 0.44 | Slight to moderate |
CU13 | B | 1.29 (0.04) | W | 0.76 (0.03) | 0.00 | 0.53 | Slight to moderate |
RA13 | C | −0.39 (0.07) | B | 0.06 (0.04) | 0.00 | −0.45 | Slight to moderate |
RA13 | C | −0.39 (0.07) | W | 0.13 (0.03) | 0.00 | −0.51 | Slight to moderate |
AA02 | B | 1.44 (0.07) | A/I | 1.02 (0.09) | 0.00 | 0.42 | Slight to moderate |
AA02 | B | 1.44 (0.07) | C | 0.96 (0.07) | 0.00 | 0.48 | Slight to moderate |
AA02 | B | 1.44 (0.07) | W | 0.82 (0.03) | 0.00 | 0.62 | Slight to moderate |
AA09 | B | 0.08 (0.03) | W | 0.50 (0.03) | 0.00 | −0.42 | Slight to moderate |
CM02 | A/I | −2.46 (0.20) | C | −1.98 (0.12) | 0.04 | −0.48 | Slight to moderate |
CM02 | A/I | −2.46 (0.20) | W | −1.73 (0.06) | 0.00 | −0.73 | Moderate to large |
CM02 | B | −2.58 (0.11) | C | −1.98 (0.12) | 0.00 | −0.60 | Slight to moderate |
CM02 | B | −2.58 (0.11) | W | −1.73 (0.06) | 0.00 | −0.85 | Moderate to large |
DIF, differential item functioning; S.E., standard error; AA, Ambiguity Acceptance; AJ, Adjustment; CM, Competitiveness; CN, Conscientiousness; CU, Curiosity; RA, Risk Approach; A/I, Asian and/or Indian; B, black African; C, mixed race; W, white.
Four instances where DIF may be present in the
Nine instances of possible DIF were identified in the
Two DIF instances were revealed across one
The potential presence of DIF was then evaluated between first-language groups.
Differential item functioning between language groups.
Item | Group 1 | Mean (S.E.) | Group 2 | Mean (S.E.) | DIF contrast | Effect size | |
---|---|---|---|---|---|---|---|
CN02 | Eng | −0.44 (0.06) | AL | −0.04 (0.06) | 0.00 | −0.41 | Approaching slight |
CN04 | Afr | −0.93 (0.09) | AL | −1.42 (0.13) | 0.00 | 0.48 | Slight to moderate |
CN04 | Eng | −0.93 (0.07) | AL | −1.42 (0.13) | 0.00 | 0.48 | Slight to moderate |
CU02 | Afr | −0.37 (0.06) | AL | −0.88 (0.08) | 0.00 | 0.51 | Slight to moderate |
CU04 | Afr | −0.83 (0.07) | AL | −1.30 (0.10) | 0.00 | 0.48 | Slight to moderate |
CU04 | Eng | −0.84 (0.06) | AL | −1.30 (0.10) | 0.00 | 0.47 | Slight to moderate |
CU13 | Afr | 0.87 (0.04) | AL | 1.32 (0.04) | 0.00 | −0.45 | Slight to moderate |
CU13 | Eng | 0.75 (0.03) | AL | 1.32 (0.04) | 0.00 | −0.57 | Slight to moderate |
AA02 | Afr | 0.92 (0.04) | AL | 1.44 (0.07) | 0.00 | −0.52 | Slight to moderate |
AA02 | Eng | 0.87 (0.04) | AL | 1.44 (0.07) | 0.00 | −0.57 | Slight to moderate |
CM02 | Afr | −1.80 (0.08) | AL | −2.51 (0.12) | 0.00 | 0.71 | Moderate to large |
CM02 | Eng | −1.98 (0.07) | AL | −2.51 (0.12) | 0.00 | 0.53 | Slight to moderate |
DIF, differential item functioning; S.E., standard error; AA, Ambiguity Acceptance; AJ, Adjustment; CM, Competitiveness; CN, Conscientiousness; CU, Curiosity; RA, Risk Approach; AL, African Languages; Afr, Afrikaans; Eng, English.
Three instances of DIF were detected across two
One item had been identified with two instances of potential DIF in the
The
This study set out to evaluate the psychometric properties of the HPTi, a personality assessment. The psychometric properties were evaluated through Rasch analysis, namely person reliability and separation, fit to the Rasch model and the Rasch version of DIF.
The reliability indices of five of the six HPTi scales would not be considered reliable against the widely accepted minimum standards. The five scales are
Item fit statistics evaluated how well the items of each HPTi trait conformed to the Rasch model. Fit mean-squared statistics greater than 1.40 indicate that the item is 40% less predictable (more varied) than the model. The same statistic under 0.60 indicates that the item is 40% more predictable (less varied) than the model expects (Bond et al.,
The DIF is an evaluation of how congruently the items of a measure define a construct between certain groups (Boone, et al.,
Across all scales, while statistically significant DIF was found between gender groups, the findings were not practically significant, as measured by DIF contrast (Linacre,
Apart from the
This study is not without limitations. Firstly, with respect to the methodology, is the use of secondary data obtained from an organisation whose use in psychometric tools is largely in the recruitment sector (thomas.co, n.d.). This falls into the disadvantage of secondary data research expressed by Boslaugh (
The study aimed to evaluate the psychometric properties of the six personality subscales of the HPTi through Rasch analysis and in the South African context. The core properties in question were reliability, fit to the Rasch model and DIF. The results indicate that while some subscales have some redeeming qualities, all subscales have their shortcomings and could be improved on for use in the South African environment. Similar shortcomings have been acknowledged historically and found in more recent evaluations of personality instruments in South Africa. Two other personality instruments, using similar evaluation techniques, achieved high reliability and good fit to the Rasch model but still experienced DIF in certain items between either ethnicity or home language (Grobler & De Beer,
The authors would like to acknowledge and appreciate Stephen Cuppello at Thomas International Ltd for granting permission to use the organisation’s data, as well as providing the data for the authors to utilise in this study.
The authors declare that, at the time of submission, D.S.S. was employed by the intellectual property rights holders, Thomas International Ltd.
D.S.S. was responsible for conceptualisation, methodology, formal analysis, data curation, resources and writing the original draft.
D.J.F.M. was the supervisor of the study and contributed toward the conceptualisation and the reviewing and editing of the manuscript.
The authors have declared that, with the tutelage and reviewership of D.J.F.M., the primary contribution of this article is credited to D.S.S.
This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.
The data that support the findings of this study are available from the corresponding author, upon reasonable request.
The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of any affiliated agency of the authors.