Does the single-item self-rated health measure the same thing across different wordings? Construct validity study

Area of Science:

• Health Services Research
• Psychometrics
• Survey Methodology

Background:

• Self-rated health (SRH) is a key indicator in health surveys.
• Variations in SRH question wording and response options can impact comparability.
• Standardizing SRH measurement is crucial for accurate health trend analysis.

Purpose of the Study:

• To assess the equivalence of three different self-rated health (SRH) forms.
• To identify the SRH form with the strongest construct validity.
• To determine the optimal coding scheme for maximizing SRH form equivalence.

Main Methods:

• Analysis of data from 58,023 respondents in the Swiss Health Survey.
• Comparison of three distinct SRH forms with varying response options.
• Assessment of construct validity using 34 health predictors and exploration of four coding schemes (continuous and dichotomous).

Main Results:

• Response option distributions varied, particularly for 'good' and 'very good' categories across forms.
• Form 3 demonstrated the highest overall explained variance, indicating superior construct validity.
• A linear coding scheme significantly enhanced the equivalence between different SRH forms.

Conclusions:

• The three self-rated health (SRH) forms examined were not fully equivalent in terms of construct validity.
• For longitudinal or cross-survey comparisons of SRH, employing a linear coding scheme is recommended to improve measurement equivalence.
• Future health surveys should consider these findings to ensure consistent SRH data collection and analysis.