I am planning a study were I need to compare between questionnaire and accelerometer data, so basically I was thinking about pearson's and spearman's correlations (if you have another idea anyway I will be happy to try it, so feel free to say it) I think that a spearman correlation would be more suitable, but I was also thinking that I could use a linear regression to use the pearson's at the end I think I will use both but I would like to hear what people who understand statistic better than me think about it.

  • $\begingroup$ If you have both questionnaire data and accelerometer data for a single set of individuals then something interesting would be to use Multiple Factor Analysis. It would allow you to compare your individuals on the basis of how they are described by each dataset. Plus you can directly compare each variable group. Note that this method only makes sens if same individuals are described by both datasets. $\endgroup$ – Riff Jun 8 '16 at 9:52
  • $\begingroup$ showing some sample data would help $\endgroup$ – Antoine Jun 8 '16 at 9:54
  • $\begingroup$ @antonie, when i will have them i will post same samples, hope to have them soon / @ nicolas, thanks for your answer this MFA seems interesting! I will have data from both the measurements for the same individuals, the aim of this study is understand if the questionnaire is good or not $\endgroup$ – aster94 Jun 8 '16 at 12:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.