Misc. Stats Topics Discussion > Training effectiveness evaluation (pretest, posttest analysis)
Thank you for the kind words, Kanan. Unfortunately, I don't have attachment functionality built into the forum. The forum is mainly intended for answering specific and concise questions, rather than guiding projects on a more broad/complex level. If you have specific questions, you are welcome to try to explain them in a post and I'll try my best to respond. If you feel a forum post won't adequately demonstrate the help that you need, that is likely a sign that the forum is not the appropriate place to receive that help.
If you're looking for more in-depth assistance, you're welcome to contact me about the paid consulting services that I also offer. Alternatively, you could try telling us the topic you're looking for assistance with and I can try to suggest a book or other resource as well.
Jeremy Taylor
Dear Mr Jeremy,
I had given up on hearing from you. Thanks for the reply anyway. The doubt I had pertains to a statistical method attributed to one Sandra J Merwin to evaluate training effectiveness by analysing pre-test and post-test scores of students. Apparently, a modified version of the paired t-test technique has been used to interpret the results. The little information that I have been able to gather from the net/books on t-tests suggests that we first need to form hypotheses and specify a level of significance. I do not understand how these conditions have been met in the said method and how the "index of learning" has been quantified. It is a very short description listed at Pg 251 of a pdf titled "A BACKGROUND FOR USING LECTURETTES IN HUMAN RESOURCE DEVELOPMENT " which shows up if googled. I would be grateful for your views on the subject but would fully understand if you chose to ignore the request. Regards. Kanan Kumar
Kanan Kumar
Dear Mr Jeremy,
1. I did some research and narrowed down the scope of my problem. The same is stated as follows.
2. The clarification/ help sought is with reference to an article on the web on "evaluation of training effectiveness". The author suggests that the "t-value" obtained from a "Paired t-test" conducted using pre-test and post test scores can be used to quantify the effectiveness of training.
3. The t-value was termed as the "Index of Learning" and the author claimed the following:-
(a) An "Index of learning" (t-value) between 0 to 1.5 indicates no evidence of learning.
(b) A t-value between 1.5 to 2.0 indicates some evidence of learning.
(c) A t-value between 2 to 3 indicates strong evidence of learning.
(d) A t-value above 3 indicates very strong strong evidence of learning.
4. In my limited understanding of the subject, the t-value, critical t-value and p-value of a t-test would depend on the significance level and degrees of freedom of the associated data. I have thus not understood the rationale behind the generalization made by the author (Para 3 (a) to 3 (c) above). I would be extremely grateful for any explanations. Thanks.
Kanan Kumar
Hey Kanan Kumar,
I'm afraid your question sounds like it is highly specific to your area of study, so I'm not familiar with the Index your are referencing. You are absolutely correct in your assessment that typically a " t-test would depend on the significance level and degrees of freedom", so the standards of reference you list are not something I'm at all familiar with. I apologize that I can't be of more assistance. If you want to see if anyone can review the topic in more depth, you are welcome to post a link to the article in a response and see if I or anyone else can review it and offer any insight, but I certainly can't make any promises about that.
Jeremy Taylor
Dear Mr Jeremy and fellow statisticians,
1. Happened to land at this website by sheer chance. Saw some of the posts and the user friendly manner followed to explain difficult concepts. Great job Mr Jeremy – do keep up the good work!
2. I need your help on the subject indicated but need to attach a small word/PDF file to indicate the details. Is there a way out?