d4

What it does

The skill focuses on developing measurement instruments (scales, questionnaires, surveys) for social science research, covering item development, validity evidence, and reliability testing.

How it works

• It provides structured stages for survey item development, including item writing guidelines (clarity, neutrality, specificity, and avoidances) and response format design (likert_scale, semantic_differential, visual_analog, forced_choice, ranking, checklist).
• It outlines a multi-stage scale construction process: stage_1 conceptualization, stage_2 item_generation, stage_3 expert_review, stage_4 cognitive_interview, stage_5 pilot_test, stage_6_validation_study, each with activities, durations, datasets, and outputs.
• It details a validity evidence framework aligned with five sources: content, response processes, internal structure, relations with other variables, and consequences, each with methods and documentation requirements.
• It specifies a reliability assessment module focusing on internal_consistency (cronbach_alpha, omega) and associated interpretations, calculations, and limitations.

When to use it

• Use when you need to develop a psychometrically sound measurement instrument for social science constructs, from item generation through validation and reliability testing, to establish a final scale and scoring instructions.

What it can touch

• Tools and scripts mentioned: SPSS (for reliability analysis) and R (psych::omega()) are implied for computations in reliability assessment.
• Methods include item-total correlations, factor analysis (EFA/CFA), Cronbach's alpha, omega, and model fit indices, as part of the validation workflow.

Caveats

• The material describes thresholds and criteria (e.g., alpha thresholds, CVR rules, fit indices) as guidelines; applicability depends on study design and context.
• It requires multiple data collection stages (pilot, validation study) with specified sample sizes and durations, which may affect project timelines.