I have a dataset with unique observations for each independent variable and observations at a grouped level for each dependent variable. What is the best model to asses if these two variable are related? I know a hierarchical/multilevel model only works when the dependent variable is at the lowest level. An OLS seems to work, except the residuals are in no way normally distributed. This almost looks like a type of ANOVA would be appropriate. I want to see if there is a statistically significant relationship between the variables, not make predictions given new observations. Here is the data:
MUAC BIODIVERSITY
1 -0.82795699 3.444444
2 -0.19090909 3.444444
3 0.34146341 3.444444
4 -2.26470588 3.444444
5 -1.37254902 3.444444
6 -1.80412371 3.444444
7 -1.48484849 3.444444
8 -1.16666667 3.444444
9 -0.82795699 3.444444
10 -1.40000000 3.444444
11 -0.73684211 3.444444
...
95 -1.51546392 2.937500
96 -3.00000000 2.937500
97 -1.85294118 2.937500
98 -1.37254902 2.937500
99 -1.97272727 2.937500
100 -1.46913580 2.937500
101 -1.86666667 2.937500
102 -1.43010753 2.937500
103 -1.04597701 2.937500
104 -1.08247423 2.937500
105 0.13725490 2.937500
...
211 -0.96078431 3.818182
212 -1.52727273 3.818182
213 -0.97849462 3.818182
214 -0.38181818 3.818182
215 -0.95454545 3.818182
216 -0.38181818 3.818182
217 -0.25925926 3.818182
218 0.08641975 3.818182
219 -0.38181818 3.818182
220 -0.63636364 3.818182
221 -1.17894737 3.818182
Is there any way to relate MUAC and BIODIVERSITY, with BIODIVERSITY as the outcome variable?
