Severe ME Day of Understanding and Remembrance: Aug. 8, 2017
Determined to paper the Internet with articles about ME, Jody Smith brings some additional focus to Severe Myalgic Encephalomyelitis Day of Understanding and Remembrance on Aug. 8, 2017 ...
Discuss the article on the Forums.

Comparison of methods for scoring & statistical analysis of SF-36 health profile & summary measures

Discussion in 'Other Health News and Research' started by Dolphin, Oct 10, 2013.

  1. Dolphin

    Dolphin Senior Member

    This is likely to be a minority interest. It relates to the SF-36 questionnaire; this is used a lot in ME/CFS research so I think a few people e.g. those who have been interested in discussing various PACE Trial papers, may find it of interest.

    However, I don't intend to write a big summary.

    Valentijn and Simon like this.
  2. Dolphin

    Dolphin Senior Member

    Here is Table 5.

    Most of these scores are relative validity scores (all except the Manova F row). If there is a gap, it means the result wasn't statistically significant. The higher the relative validity co-effecient, the better the measure is.

    The acronyms are the usual for the SF-36
    PF = Physical Functioning subscale
    RP = Role Physical subscale
    BP = Bodily Pain subscale
    GH = General Health subscale
    VT = Vitality subscale
    SF = Social Functioniong subscale
    RE = Role Emotional subscale
    MH = Mental Health subscale
    PCS = Physical Component Score (a combination of the subscale scores)
    MCS = Mental Component Score (a different combination of the subscale scores)


    So what this shows the subscale which is most valid for a conditions/symptom cluster varies.

    An F score checks for variations between groups. The way I recall understanding it is if one had a certain number of trees (or sectors) and things grew around them, one would could check to see whether certain sectors were better or worse for growth (to take one measurement). Some sectors might have better soil, light, water, etc. So it checks for variations.

    There are other tables in the paper that are also interesting, but I don't have time at the moment to do a proper summary.
    Valentijn likes this.

See more popular forum discussions.

Share This Page