Data analysis

Ahyanvirg

New Member
Hi. need help. I am new to research data analysis and the last time I came across it was when I was still in Uni. To cut the story short, wanted some advise on the research analysis that I am doing. We wanted to know what scores from candidate's pre-employment data can predict a high scores in Training outcome (probability of candidate to excel or to be dismissed from training). Which statistical analysis can we use?

Karabiner

TS Contributor
What data from pre-employment do you have at hand?

What is the dependent variable, the training score (e.g. something like a score ranging from 0 to 20)?
Or is it categorical (dismissed/excellent/normal) ?

Very important: how large is your sample size?

Also important: what do you need the results for. Do you want to analyse the past data just out of interest,
or do you also want to use the results for something in the future (if yes, what for)?

With kind regards

Karabiner

Ahyanvirg

New Member
What data from pre-employment do you have at hand?

What is the dependent variable, the training score (e.g. something like a score ranging from 0 to 20)?
Or is it categorical (dismissed/excellent/normal) ?

Very important: how large is your sample size?

Also important: what do you need the results for. Do you want to analyse the past data just out of interest,
or do you also want to use the results for something in the future (if yes, what for)?

With kind regards

Karabiner

I have an aptitude test scores from pre-employment data (scores are ranging from 15-45) and I have 5 different training data outcome, each factor has scores ranging from 1-5.

I want to know if the pre-employment test can predict the outcome of that 5 training test.

I have 433 samples

Karabiner

TS Contributor
If I understand you correctely, you can sum up the 5 training outcomes in order to achieve a scale ranging from 5 to 25? If yes, then I would start by creating an x-y scatterplot with aptitude test and the training data sum score (so that you can get a visual impression of how the variables are associated), and by calculating a Pearson correlation coefficient and a Spearman rank correlation coefficient.

With kind regards

Karabiner

Ahyanvirg

New Member
If I understand you correctely, you can sum up the 5 training outcomes in order to achieve a scale ranging from 5 to 25? If yes, then I would start by creating an x-y scatterplot with aptitude test and the training data sum score (so that you can get a visual impression of how the variables are associated), and by calculating a Pearson correlation coefficient and a Spearman rank correlation coefficient.

With kind regards

Karabiner
Hi again. What I meant on the 5 Training outcome is that there are 5 factors being tested during training, each of them are scored from 1-5 (e.g. "Confidence, Handling, Knowledge...") I want to know if the aptitude test score can predict or the correlation of it to each of these 5 factor scores.

Karabiner

TS Contributor
So you can create 5 scatterplots and 5 Spearman rank correlations (since it is safe
to consider the 5 point scale as an ordinal scale, and a rank correlation is adequate
for this).

With kind regards

Karabiner

Ahyanvirg

New Member
So you can create 5 scatterplots and 5 Spearman rank correlations (since it is safe
to consider the 5 point scale as an ordinal scale, and a rank correlation is adequate
for this).

With kind regards

Karabiner
Thank you so much for the help!