Hi everyone, I've come across something a little strange in a study by Baum (2014), measuring the effects of a 5-day sleep restricted condition on adolescent mood using the Profile of Mood States (POMS).
Analysis is run through Wilcoxon Signed Rank Tests, comparing a sleep restricted condition (6.5 hours time in bed) to a healthy sleep condition (10 hours time in bed).
I'm looking for an effect size to report, but all that is reported is:
Compared with HS, during SR, adolescents rated themselves as having increased levels of Tension/Anxiety (Z = -3.25, p = .001), Anger/Hostility (Z = -3.10, p = .002), Fatigue/Inertia (Z = -4.78, p < .001), and Confusion/Bewilderment (Z = -3.37, p = .001), and significantly less vigor (Z = -3.02, p = .003).
I've found a little bit of information on Z-factors which I believe I'm seeing here, but I have no idea how to interpret the effect size (small, medium, large?)
Can someone please shed some light?
Thank you.
z is a standard score for the T distribution: z = (T minus mean T) / std dev of T. See page 79 in Siegel's 'Nonparametric Statistics for the Behavioral Sciences' (McGraw-Hill, 1956 or a newer edition) in the section on the Wilcoxon Test.
The above answer is a correct analysis of Z scores. When using the Profile of Mood states as your outcome marker,. You need to ignore the Z scores and work with raw scores computing a r squared value comparing experimental participants and controls on raw data to determine the effect size of your intervention. We have found the effect size is small or large using the POMS depending on biochemical assays of blood concentrations of the water soluble vitamins.
Dear Stephen, I think you want Rosenthal's formula to do this: r=Z/√N. See this paper, which describes the formula and gives the reference we used, as well as how to interpret the resulting effect size. Good luck.
Article Ventilatory function in children with severe motor disorders...
- Badges
- Science topic
- Similar topics
- Psychology
- Developmental Psychology