Article Text

Download PDFPDF
Do different scales measure the same construct? Three Sense of Coherence scales
  1. M Olsson1,
  2. J Gassne2,
  3. K Hansson2
  1. 1
    The Danish National Institute of Social Research, København, Denmark
  2. 2
    School of Social Work, Lund University, Sweden
  1. Dr M Olsson, The Danish National Institute of Social Research, Herluf Trolles Gade 11, DK-1052 København K, Denmark; mao{at}


Background: Different scales claim to measure the construct “Sense of Coherence”. Results from these scales have been compared without knowing whether they measure the same construct. This article compares two versions of Antonovsky’s original scale (SOC-13 and SOC-29), translated into Swedish, and a three-item scale (SOC-3) that claims to measure Sense of Coherence.

Methods: The data were analysed in a cross-sectional setting. The study consisted of university students studying social work (n = 395).

Results: The original scales had no distribution problems in differentiating Sense of Coherence. The SOC-3 had severe distribution problems. The two versions of the original Sense of Coherence scale had an acceptable reliability (Cronbach’s α; SOC-29 = 0.93, SOC-13 = 0.89). The SOC-3 scale did not have an acceptable reliability (Cronbach’s α = 0.39). SOC-29 and SOC-13 had a high intercorrelation (r = 0.96, p<0.001). The SOC-3 significantly correlated with SOC-29 (r = −0.72, p<0.001) and SOC-13 (r = −0.67, p<0.001), but the magnitude was significantly lower than the intercorrelation between SOC-29 and SOC-13 (Fisher’s z-transformation, p<0.001).

Conclusions: Because scales that claim to measure the same construct are not always interchangeable, researchers should make sure they compare results from studies that use the same scales.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


  • Competing interests: None declared.