1
$\begingroup$

How to interpret my chart? I want to get the maximum likelihood in logistic regression with this result (I'm really not sure if this is how it looks like): enter image description here

I am currently using logistic regression to National Achievement Test(a performance exam for students,NAT -GRADE-REMARKS the Y axis) and their scholastic grade(In the example above 2CLPW(21st Century Literature in the Philippines and the World)-(Grade 12)-Q1(Quarter 1), the x Axis).I am using Rapidminer.

I wanted to know the likelihood of students to Pass the National Achievement Test or to get VLM or LM. For my example in the image above the category in the National Achievement Test is VLM(Very Low Mastery) which is set to 1 and the LM(Low Master) Set to 0. VLM and LM are the only categories all the student gets.

I wanted to know the maximum likelihood in this graph in order to fit a S like line (sigmoid function).Can I fit a sigmoid S shape? If there are no relationship then that means I can't fit an S shape line right? I just really dont know how to interpret the scatterplot above. Do I need to zoom it out? I really cant interpret it.Green is Very Low Mastery, Blue is Low Mastery

$\endgroup$
4
  • 1
    $\begingroup$ In what way is logistic regression insufficient for your problem? $\endgroup$ – Ben Reiniger Sep 6 '20 at 1:08
  • $\begingroup$ @BenReiniger its not insufficient, I just don't know how to interpret my chart whether they have relationship $\endgroup$ – Nathalie Joy Galia Sep 6 '20 at 2:44
  • $\begingroup$ Plot the Ratios rather than Qty and use a polynomial fit with a high R^2 $\endgroup$ – Tony Stewart Sunnyskyguy EE75 Sep 6 '20 at 5:20
  • $\begingroup$ What doyou mean by that? I just wanted to know the interpretation of my scatter plot with logistic regression. $\endgroup$ – Nathalie Joy Galia Sep 6 '20 at 8:50