Table 2.
Summary of current practices of the scale development process
| Methods | Number of scales resorting to | Percentage (%) of scales resorting to |
|---|---|---|
| Step 1—item generation | ||
| Deductive methods (exclusively) | 37 | 35.2 |
| Inductive methods (exclusively) | 8 | 7.6 |
| Combined deductive and inductive methods | 59 | 56.2 |
| Literature review | 89 | 84.7 |
| Existing scales | 40 | 38 |
| Interviews | 28 | 26.6 |
| Focus groups | 25 | 23.8 |
| Expert panel | 23 | 21.9 |
| Qualitative exploratory research | 3 | 5 |
| Not clearly reported method | 1 | 1 |
| Step 2—theoretical analysis | ||
| Expert judges | 78 | 74.2 |
| Target population judges | 46 | 43.8 |
| Use of just one approach | 67 | 63.8 |
| Combined two approaches | 29 | 27.7 |
| Not clearly reported approach | 9 | 8.5 |
| Step 3—psychometric analysis | ||
| EFA | 93 | 88.6 |
| CFA | 76 | 72.3 |
| Combined EFA and CFA | 69 | 65.7 |
| Lack of EFA and CFA | 5 | 4.7 |
| Convergent/concurrent validity | 76 | 72.3 |
| Discriminant validity | 59 | 56.2 |
| Predictive/nomological validity | 34 | 32.3 |
| Criterion validity | 17 | 16.2 |
| External validity | 5 | 4.7 |
| Internal validity | 3 | 2.8 |
| Internal consistency | 105 | 100 |
| Test-retest reliability | 24 | 22.8 |
| Item-total correlation/inter-item reliability | 19 | 18.1 |
| Split-half reliability | 3 | 2.9 |
| Inter-judge reliability | 3 | 2.9 |
| Sample size about step 3 and number of items | ||
| Sample size smaller than the rule of thumb 10:1 | 53 | 50.4 |
| Number of items final scale reduced by 50% | 42 | 40 |
| Number of items final scale reduced more than 50% | 52 | 49.6 |
| Not clearly reported inicial item number | 11 | 10.4 |
EFA exploratory factor analysis, CFA confirmatory factor analysis