Just making it easier to understand?
From the Journal of Nutritional Science
Young adult males (n 35) were supplemented with either half or two kiwifruit/d for 6 weeks. Profile of Mood States questionnaires were completed at baseline and following the intervention. No effect on overall mood was observed in the half a kiwifruit/d group; however, a 35 % (P = 0·06) trend towards a decrease in total mood disturbance and a 32 % (P = 0·063) trend towards a decrease in depression were observed in the two kiwifruit/d group. Subgroup analysis indicated that participants with higher baseline mood disturbance exhibited a significant 38 % (P = 0·029) decrease in total mood disturbance, as well as a 38 % (P = 0·048) decrease in fatigue, 31 % (P = 0·024) increase in vigour and a 34 % (P = 0·075) trend towards a decrease in depression, following supplementation with two kiwifruit/d. There was no effect of two kiwifruit/d on the mood scores of participants with lower baseline mood disturbance
From the Otago press release
Eating two kiwifruit a day can improve a person’s mood and give them extra energy, new research from the University of Otago, Christchurch (UOC) shows.
Over a six-week period, normally-healthy young men either ate two kiwifruit a day or half a kiwifruit daily as part of a research study into the potential mood-enhancing effects of the fruit.
Researchers found those eating two kiwifruit daily experienced significantly less fatigue and depression than the other group. They also felt they had more energy. These changes appeared to be related to the optimising of vitamin C intake with the two kiwifruit dose
Eating two kiwifruit a day can improve mood and energy levels, a new University of Otago study shows.
Those eating two kiwifruit were found to experience significantly less fatigue and depression than the others. They also felt they had more energy.
I’m not criticizing the research, which was a perfectly reasonable designed experiment, but if the findings are newsworthy, they are also worth presenting accurately.
Thomas Lumley (@tslumley) is Professor of Biostatistics at the University of Auckland. His research interests include semiparametric models, survey sampling, statistical computing, foundations of statistics, and whatever methodological problems his medical collaborators come up with. He also blogs at Biased and Inefficient See all posts by Thomas Lumley »
What’s the difference? I can’t see it.
Also, those are some weak sauce p-values.
11 years ago
The press release and story report a subgroup analysis as if it applied to everyone.
As you say, those are unimpressive p-values, especially when you consider that it’s really the wrong analysis — those are tests for change within each treatment group, not tests for difference in change between the treatment groups.
11 years ago