, PYC4807
ASSIGNMENT 4 (EXAM
PORTFOLIO) 2026
DUE 5 OCTOBER 2026
QUESTION 1: RELIABILITY OF THE UWES-S
1(a) Type of Reliability
The reliability of the Utrecht Work Engagement Scale-Student (UWES-S) is predominantly evaluated
through internal consistency and test-retest reliability, which are the two primary psychometric
properties reported in the validation literature (Schaufeli & Bakker, 2003).
Internal consistency refers to the degree of intercorrelation among items within a scale, most
commonly quantified using Cronbach’s alpha coefficient. For the UWES-S, internal consistency has
been examined across its three subscales—vigour, dedication, and absorption—as well as for the
total scale score. The generally accepted threshold for established scales is α ≥ .70, while a more
lenient criterion of α ≥ .60 applies to newly developed instruments (Nunnally & Bernstein, 1994).
Test-retest reliability, alternatively termed stability, assesses the consistency of scores over time
when the same instrument is administered to the same respondents on separate occasions. This
aspect of reliability provides evidence that the construct being measured remains stable across
time intervals and is not unduly influenced by transient situational factors.
More recent psychometric evaluations have additionally employed composite reliability and
McDonald’s omega as alternatives to Cronbach’s alpha, addressing some of the limitations
associated with the tau-equivalence assumption underlying alpha (Deng et al., 2025). These
modern approaches offer more accurate estimates, particularly when the scale demonstrates a
multifactorial or hierarchical structure.
1(b) Findings and Interpretation
The internal consistency findings for the UWES-S demonstrate generally robust reliability across
diverse samples and cultural contexts. In the original validation data, the UWES-S-9 (short version)
yielded Cronbach’s alpha values of .73 for vigour, .76 for dedication, and .70 for absorption, with a
total scale alpha of .84 (Schaufeli et al., 2006). The longer UWES-S-17 version exhibited even higher
internal consistency, with alpha coefficients of .82 for vigour, .89 for dedication, and .83 for
absorption, and a total scale alpha of .93 (Schaufeli & Bakker, 2003). These values consistently
exceed the conventional .70 threshold, indicating that the items within each subscale are
homogeneous and measure the intended construct with minimal measurement error.
, Cross-cultural validation studies have largely corroborated these findings. A Chinese validation
among early childhood pre-service teachers reported Cronbach’s alpha values of .88 for vigour, .89
for dedication, and .91 for absorption for the UWES-S-9, with the total scale reaching .95 (Li et al.,
2024). Similarly, a Korean adaptation of the 17-item version produced alpha coefficients ranging
from .83 to .94, with additional composite reliability and McDonald’s omega values of .82 to .94,
thereby reinforcing the instrument’s reliability in non-Western educational settings (Kim & Lee,
2022). In the South African context, a large-scale study involving 2,434 university students
confirmed high internal consistency for the UWES-S, supporting its reliable application across
language and campus groups (Brand et al., 2025).
A notable exception to this pattern emerged in a South African sample where the vigour subscale
of the UWES-15 yielded an alpha of .65, falling below the .70 threshold, while absorption
yielded .61 (Van der Walt, 2014). The researcher identified problematic items—specifically item six
for vigour and item seventeen for absorption—that negatively affected reliability, leading to their
removal and the adoption of a revised 15-item version. Following these modifications, all subscale
alphas exceeded .80, and the total scale alpha reached .96, demonstrating that item-level
refinement can substantially enhance reliability.
Regarding test-retest reliability, a 14-day interval study with 203 Chinese undergraduate students
reported an intraclass correlation coefficient (ICC) of .771 for the overall UWES-S-9, with subscale
ICCs of .745 for vigour, .778 for dedication, and .773 for absorption (Deng et al., 2025). These
coefficients fall within the moderate to good range, indicating that the instrument possesses
acceptable temporal stability. However, it should be noted that reliability coefficients, whether
internal consistency or test-retest, are sample-dependent and may vary across populations,
educational levels, and cultural contexts.
The interpretation of these reliability findings must also consider recent bifactor modelling
evidence suggesting that the UWES-S may be predominantly unidimensional, with a strong general
engagement factor accounting for the majority of reliable variance (Deng et al., 2025). This raises
questions about the incremental utility of subscale scores, as omega hierarchical coefficients for
specific factors were found to be minimal.
1(c) Implications
The consistently high internal consistency of the UWES-S has several important implications for
both research and practice. For researchers, the reliability evidence supports the use of the
instrument to measure academic engagement as a multidimensional construct, with subscale
scores providing meaningful differentiation between vigour, dedication, and absorption. The strong
total scale reliability (α ≥ .84 across versions) further justifies the use of composite engagement
scores in studies where a global index is theoretically appropriate (Schaufeli & Bakker, 2003). This
dual capacity—to assess both specific dimensions and overall engagement—enhances the
instrument’s versatility for addressing diverse research questions.
ASSIGNMENT 4 (EXAM
PORTFOLIO) 2026
DUE 5 OCTOBER 2026
QUESTION 1: RELIABILITY OF THE UWES-S
1(a) Type of Reliability
The reliability of the Utrecht Work Engagement Scale-Student (UWES-S) is predominantly evaluated
through internal consistency and test-retest reliability, which are the two primary psychometric
properties reported in the validation literature (Schaufeli & Bakker, 2003).
Internal consistency refers to the degree of intercorrelation among items within a scale, most
commonly quantified using Cronbach’s alpha coefficient. For the UWES-S, internal consistency has
been examined across its three subscales—vigour, dedication, and absorption—as well as for the
total scale score. The generally accepted threshold for established scales is α ≥ .70, while a more
lenient criterion of α ≥ .60 applies to newly developed instruments (Nunnally & Bernstein, 1994).
Test-retest reliability, alternatively termed stability, assesses the consistency of scores over time
when the same instrument is administered to the same respondents on separate occasions. This
aspect of reliability provides evidence that the construct being measured remains stable across
time intervals and is not unduly influenced by transient situational factors.
More recent psychometric evaluations have additionally employed composite reliability and
McDonald’s omega as alternatives to Cronbach’s alpha, addressing some of the limitations
associated with the tau-equivalence assumption underlying alpha (Deng et al., 2025). These
modern approaches offer more accurate estimates, particularly when the scale demonstrates a
multifactorial or hierarchical structure.
1(b) Findings and Interpretation
The internal consistency findings for the UWES-S demonstrate generally robust reliability across
diverse samples and cultural contexts. In the original validation data, the UWES-S-9 (short version)
yielded Cronbach’s alpha values of .73 for vigour, .76 for dedication, and .70 for absorption, with a
total scale alpha of .84 (Schaufeli et al., 2006). The longer UWES-S-17 version exhibited even higher
internal consistency, with alpha coefficients of .82 for vigour, .89 for dedication, and .83 for
absorption, and a total scale alpha of .93 (Schaufeli & Bakker, 2003). These values consistently
exceed the conventional .70 threshold, indicating that the items within each subscale are
homogeneous and measure the intended construct with minimal measurement error.
, Cross-cultural validation studies have largely corroborated these findings. A Chinese validation
among early childhood pre-service teachers reported Cronbach’s alpha values of .88 for vigour, .89
for dedication, and .91 for absorption for the UWES-S-9, with the total scale reaching .95 (Li et al.,
2024). Similarly, a Korean adaptation of the 17-item version produced alpha coefficients ranging
from .83 to .94, with additional composite reliability and McDonald’s omega values of .82 to .94,
thereby reinforcing the instrument’s reliability in non-Western educational settings (Kim & Lee,
2022). In the South African context, a large-scale study involving 2,434 university students
confirmed high internal consistency for the UWES-S, supporting its reliable application across
language and campus groups (Brand et al., 2025).
A notable exception to this pattern emerged in a South African sample where the vigour subscale
of the UWES-15 yielded an alpha of .65, falling below the .70 threshold, while absorption
yielded .61 (Van der Walt, 2014). The researcher identified problematic items—specifically item six
for vigour and item seventeen for absorption—that negatively affected reliability, leading to their
removal and the adoption of a revised 15-item version. Following these modifications, all subscale
alphas exceeded .80, and the total scale alpha reached .96, demonstrating that item-level
refinement can substantially enhance reliability.
Regarding test-retest reliability, a 14-day interval study with 203 Chinese undergraduate students
reported an intraclass correlation coefficient (ICC) of .771 for the overall UWES-S-9, with subscale
ICCs of .745 for vigour, .778 for dedication, and .773 for absorption (Deng et al., 2025). These
coefficients fall within the moderate to good range, indicating that the instrument possesses
acceptable temporal stability. However, it should be noted that reliability coefficients, whether
internal consistency or test-retest, are sample-dependent and may vary across populations,
educational levels, and cultural contexts.
The interpretation of these reliability findings must also consider recent bifactor modelling
evidence suggesting that the UWES-S may be predominantly unidimensional, with a strong general
engagement factor accounting for the majority of reliable variance (Deng et al., 2025). This raises
questions about the incremental utility of subscale scores, as omega hierarchical coefficients for
specific factors were found to be minimal.
1(c) Implications
The consistently high internal consistency of the UWES-S has several important implications for
both research and practice. For researchers, the reliability evidence supports the use of the
instrument to measure academic engagement as a multidimensional construct, with subscale
scores providing meaningful differentiation between vigour, dedication, and absorption. The strong
total scale reliability (α ≥ .84 across versions) further justifies the use of composite engagement
scores in studies where a global index is theoretically appropriate (Schaufeli & Bakker, 2003). This
dual capacity—to assess both specific dimensions and overall engagement—enhances the
instrument’s versatility for addressing diverse research questions.
