Hello,
I'm working on a research project and was wondering if anyone could point me in the right direction of analyzing data for an experiment which would have the following:
64 readings, in numerical value
Non-numerical possible correlating variables are:
Strain: Strain 1, Strain 2, and Strain 3
Humidity: Low, high
Treatment: Treatment 1, Treatment 2
I have everything labeled side by side on excel but am not sure how to start digesting it.
Thank you for any guidance you may give!
Hello David Munoz,
If your numeric response is continuous, I suggest you fit a multiple linear regression with predictors the variables you are mentioning. You can perform it in SPSS, R, SAS or any other statistical software you may prefer.
Make sure to perform an appropriate model selection procedure and check model assumptions before you arrive at conclusions.
If you work on SPSS you can find more details here:
https://statistics.laerd.com/spss-tutorials/multiple-regression-using-spss-statistics.php
If you prefer R :
https://www.r-bloggers.com/r-tutorial-series-multiple-linear-regression/
Good luck with your work!
I will suggest SPSS.
you may need to first perform some test such normality test etc before selecting the appropriate Multivraint test.
If you have numeric readings and if the variables you mentioned were applied in combinations, this looks like a factorial design set in 3 strains by 2 humidity levels by 2 treatments. But I am not sure what number of replication you used that gives 64 observations. The it is possible to carry out analysis of variance in factorial to see the main effects of each variable and their two-and three-way interactions.
You can equally used artificial intelligenece based models... It will equally help you in identifying whether the data set is non-linear or linear. Its absolutely of paramount importance in modelling non-linear models. And provides better performance skills than the linear models such as multivariate regression model, Multi linear regressions etec
- Badges
- Science topic