Table 2 Qualitatively summarized results, overall rating, and quality of evidence (GRADE) on structural validity per instrument
Instrument (ref) | Summarized result on structural validity | Rating | Quality of evidence | |
|---|---|---|---|---|
Quality level | Reason | |||
EFA and CFA. Four factors/subscales. Inconsistent results. Results for seven studies met the criteria (SRMR < 0.08), three studies did not (SRMR > 0.08/RMSEA > 0.06) | ( +) | Moderate | − 1 incon | |
EFA and CFA. Eight new factors. Inconsistent results, different number of factors in two studies. EFA criteria were met on eight-factor structure, no fit indices reported on CFA | ( +) | Low | − 1 incon, − 1 RoB | |
CFA showed a 36-item 12-factor scale. Inconsistent results. One study met the criteria (RMSEA = 0.052) and one study did not (RMSEA = 0.64) | ( ±) | No grade | ||
EFA and CFA. Three factors/subscales. Inconsistent results. Results for five studies met the criteria for CFA (SRMR < 0.08) or criteria for EFA, two studies did not | ( +) | Moderate | − 1 incon | |
EFA. Disagreement about dimensionality between five studies. Results summarized in subgroups (unidimensional/multifactorial) 1. Unidimensional (single factor): criteria for EFA were met in both (two) 2. Multifactorial (four factors): inconsistent results. Criteria not met in two studies due to cross-loading, and the third for not reporting cross-loading | 1: ( +) 2: ( −) | Moderate Moderate | − 1 RoB − 1 incon | |
EBP Beliefs- Short [102] | EFA. One factor. Factor loading > 0.70, Eigenvalue = 2.25 | ( +) | Low | − 2 RoB |
EFA. Disagreement about dimensionality between four studies. Results are summarized in subgroups (unidimensional/multifactorial) 1. Unidimensional (single factor): criteria for EFA were met in one study 2. Multifactorial (four/five/two factors): inconsistent results. Criteria not met in two studies due to cross-loading, and one study rated as indeterminate due to not reporting eigenvalue, total variance explained, or cross-loading | 1: ( +) 2: ( −) | Low Moderate | − 2 RoB − 1 incon | |
EBP Implement-Short [102] | EFA. One factor. Factor loading > 0.85, Eigenvalue = 2.46 | ( +) | Low | − 2 RoB |
Ethiopian EBP Implement [111] | EFA (two factors): Factor loadings > 0.40, Eigenvalues > 1, cross-loading, no cross loadings | ( +) | Moderate | − 1 Rob |
Al Zoubi Questionnaire [62] | IRT/Rasch: Unidimensionality: CFI, TLI, RMSEA or SRMR not reported. Local independence: items correlating > 0.3 led to removal. Monotonicity: not reported. Model fit (× 2 test): × 2 > 0.01 on three out of four subscales | (?) | No grade | |
EBPP-S [63] | CFA confirmed a three-factor scale. Model fit: CFI = .96, and RMSEA = .06 | ( +) | Low | − 2 RoB |
EBP inventory [66] | EFA. Four factors/subscales. Factor loadings > 0.30, cross-loading, tot variance explained, and eigenvalue not reported | (?) | No grade | |
EPIC [69] | Unidimensional scale. EFA: all items loaded into one single factor > 0.4, tot explained variance = 71%, < 10% cross-loading | ( +) | Moderate | − 1 RoB |
MPAS [72] | CFA. Unidimensional model, and a modified five-item model had the best fit: US: CFA: RMSEA = 0.030, CFI 0.998 Korea: CFA: RMSEA = < 0.05, CFI > 0.95 | ( +) ( +) | High High | |
EFA and CFA. Inconsistent result. One study met the criteria for EFA (factor loading > 0.3, eigenvalues > 1, and < 10% cross-loading). The other study did not meet the criteria for CFA (CFI = 0.90) | ( ±) | No grade | ||
EBPPAS-s [75] | CFA. Revised four-factor model, 37 items. Model fit: CFI = 0.90, RMSEA = 0.06 | ( −) | High | |
EFA and CFA. Three factors/subscales. Inconsistent results. One study met the criteria for EFA (factor loading > 0.3, tot explained variance = 73.01%, and < 10% cross-loading). One study did not meet the criteria for CFA (CFI = 0.91) | ( ±) | No grade | ||
EBP Capability Beliefs [79] | IRT/Rasch: Unidimensionality: CFI, TLI, RMSEA or SRMR not reported. Local independence: two small cross-loadings on the first factor; 0.23 and 0.30. Monotonicity: not reported. Model fit (× 2 test): × 2 = 42, 71 | (?) | No grade | |
HEAT [80] | EFA and CFA. Four-factor model. CFA model fit: SRMR = 0.063 | ( +) | High | |
EBP-KABQ [81] | CFA: Four-factor modified model. Model fit: CFI = 0.89 | ( −) | High | |
EFA and CFA. Three-factor model, inconsistent results. Model fit: CFI = 0.957 | ( +) | Moderate | − 1 Incon | |
HS-EBP [86] | EFA and CFA. Five factors. Model fit: CFI = 0.99, RMSEA = 0.047, SRMR = 0.067 | ( +) | High | |
EBPRS [88] | EFA and CFA. Four factors. Model fit: CFI = 0.96, TLI = 0.96, SRMR = 0.058 | ( +) | High | |
EFA and CFA. Inconsistent results. Five factors in three out of four studies. One study met the criteria for EFA (factor loading > 0.3, tot explained variance = 63%, and < 10% cross-loading), one did not report enough information to be rated, the third did not meet criteria for CFA (CFI = 0.69). Overall rating is inconsistent | ( ±) | No grade | ||
ISP-D [94] | CFA. Four-factor model. Model fit: SRMR = 0.075 | ( +) | High | |
EBNAQ [95] | PCA (EFA) with Promax rotation showed three factors explaining a total variance of 54.7%, factor loadings > 0.30 and cross-loading < 10% | ( +) | Moderate | − 1 RoB |
CFA. Four-factor model. Model fit: CFI = 0.93 and 0.82 in two studies | ( −) | High | ||
I-SABE [108] | EFA. Four factors. Tot variance explained = 52.6%, factor loading > 0.3, cross-loading > 10% | ( −) | Moderate | − 1 RoB |
Noor EBM [109] | EFA. 1. Attitude scale (five factors): factor loading = > 0.4, tot variance = 66.3%, cross-loading not reported. 2. Practice scale (two factors): factor loading = > 0.4, tot variance = 55.4%, cross-loading not reported | 1: (?) 2: (?) | 1,2: no grade | |
EBP-CBFRI [110] | CFA. Five-factor model. Model fit: RMSEA = 0.05 | ( +) | High | |