As big mask studies are still inconclusive you are unlikely to detect much impact in your sample.

Unless you run around with lateral flow test (in contrast to relying on counting symptomatic ILI), you may even have problems to detect statistically significant impact of vaccination on medium sized sample.

[I just warn you before you put lot's of effort into your study and later end up with inconclusive results that no journal would be willing to publish]

Unlike Covid, young schoolchildren are known to play a significant role in the spread of influenza. Vaccination of young schoolchildren can affect transmission within their families. See: Article Reduction in self-reported influenza-like-illness in school ...

Stephen C Schoenbaum

VE of the study that you linked is exceptionally high, like ~60% against any ILI (which is quite spectacular result as majority of ILI are not even influenza to begin with so should not be affected).

To compare apples to apples I looked other studies covering that 2016/2017 season, where they got much more plausible VE ~40% against symptomatic, swab test confirmed influenza.

USA:

https://academic.oup.com/cid/article/68/11/1798/5094815

UK:

Article End-of-season influenza vaccine effectiveness in adults and ...

(for UK this value is estimated VE after adjusting timing)

Any idea why they got such outlier result? (self-reported and small sample size or I miss some local epidemiological nuance?)

Rama Ibrahim

How are you measuring whether people have been washing their hands? (I mean due to social desirability bias people are highly likely to tell you that they have been washing them regardless of the reality)

It is certainly valid to wonder if vaccine effectiveness was truly as high as 60%, and, I believe it is impossible now to determine if the results from respondents to the questionnaire (less than 40 percent of families, I believe) are representative of the entire population. But it is also worth remembering that this is not the first paper to suggest that immunization of school children can have an effect on the occurrence of influenza in their families and even in the community. Monto’s study of immunization of school-age children in Tecumseh, Michigan, during the H3N2 pandemic of 1968 is the landmark study of that issue.

I think it is reasonable to think there may be a real and important effect and to keep a focus on school-based immunization (with careful evaluation).