Hello
I monitor the water quality for 7 days and check the physio-chemical properties of water so after 7 days I get 21 data points . So can anyone helps me to figure out my doubt that we can apply stats on 21 data points are not and which stats I will go ahead such as ANOVA, Pearson correlation and so on.
Dear Prachi,
You can use Pearson's R correlation but also I would suggest Kendal's Tau for smaller sample sizes. If you can, please apply both of them and compare the results. Spearman's rank coefficient correlation can be also useful if you have sudden spikes or regular outliers on your data.
Best
Ibrahim
Thamks Ibrahim sir for your valuable comment but sir if the variation in the data is that more means slight variation then which stats i will go for
Dear Prachi Arora
If your data is not normally distributed (highly varied across the mean), Spearman's rank coefficient is the best way to decrease the bias and increase the robustness. Moreover, usually in the social sciences, we try to de-trend the data to have accurate results.
Yes, it is possible to perform the analysis, but to use non parametric methods in the analysis
- Badges
- Science topic
- Similar topics
- Agricultural Science
- Soil Science
- Soil Analysis