Interpreting statistics and research (housework among boys and girls)
I just encountered a Swedish news service claiming that “Girls help [do housework] more at home”/“Flickor hjälper till mer hemma”.
While this article, to my mild surprise, did not make the usual partisan statements of e.g. “Girls are better at X”, it still manages to show some common problems with reading of statistics and how poor critical thinking can lead people (in particular, journalists) astray.
To quote relevant parts:
SCB har undersökt vilka hushålls-
sysslor barn i åldrarna 10-18 år
hjälper till med.
83 procent av flickorna och 79 pro-
cent av pojkarna hjälper till med hus-
hållsarbete minst en timme i veckan.
([“The bureau of statistics”] has investigated what household chores children in the age range 10–18 years help with.
83 per cent of the girls and 79 per cent of the boys help with household work for at least an hour a week.)
Syssla; Flickor; Pojkar
Bäddar sin säng; 82 proc; 77 proc
Diskar eller plockar i/ur diskmaskinen; 81 proc; 71 proc
Städar sitt rum; 78 proc; 64 proc
Tar hand om syskon; 35 proc; 36 proc
Arbetar utomhus; 23 proc; 40 proc
(Task; Girls; Boys
Makes own bed; 82 %; 77 %
Does the dishes or loads/unloads the dish-washer; 81 %; 71 %
Cleans own room; 78 %; 64 %
Takes care of siblings; 35 %; 36 %
Works outdoors; 23 %; 40 %)
(The news service in questione does not provide an archive, so I cannot give a permanent link. Should I encounter the data from another source, I will add one.)
Going by the numbers presented (but beware that the full report may give another view; for instance, the list of task is likely to be abbreviated), the claim is highly dubious. Firstly, the difference in overall numbers is comparatively small (certainly not large enough to allow for predictions about individuals) and, depending on the size of the sample, could lack statistical significance. Secondly, and more importantly, the tasks are oddly chosen:
Both making ones own bed and cleaning ones own room are things that do not constitute “helping at home”—they are something that a child in the age bracket given either does or does not do for his/her own benefit. (Similarly, baking cookies for ones own consumption is not “helping at home” either—nor is tweaking ones own moped.)
The natural step would be to adjust the overall numbers by removing these entries. For lack of in-depth data, this is not possible; however, we can make a very rough first comparison by simply adding percentages. Now, in the original version we have 82 + 81 + 78 + 35 + 23 = 299 for the girls and 77 + 71 + 64 + 36 + 40 = 288 for the boys. (Pleasingly, 288 / 299 * 83 is just shy of 80, which compares well to the original 79 % overall for boys—in particular, as rounding can cause minor distortions.) Removing the “self serving” tasks, we instead have 81 + 35 + 23 = 139 for the girls and 71 + 36 + 40 = 147 for the boys—who are now ahead by more than they used to trail (as a proportion of the total).
The tentative conclusion, then: Boys (!) help more at home. (Incidentally and anecdotally: This was definitely the case when looking at me and my sister as teenagers. She could barely be bothered to put her own plates in the dish-washer; I moved the lawn and chopped wood for the fireplace.) Of course, I cannot guarantee that this would remain true if the raw data was re-investigated, but the gap is sufficiently large that the original claim (that girls help more) should be viewed as unsupported.
As an aside, the removed categories reflect an issue that is worth keeping in mind when discussing housework: Men and women have different priorities when it comes to cleaning and use of available time. (In my opinion, men have it the right way around and women should take a more relaxed attitude.)