943 reputation
314
bio website stats.stackexchange.com/users/…
location The Earth
age
visits member for 1 year, 9 months
seen Apr 10 at 14:30

Learner.


Mar
4
awarded  Popular Question
Feb
24
awarded  Popular Question
Nov
22
awarded  Popular Question
Nov
21
awarded  Notable Question
Nov
13
comment Which test should I run?
If you want to compare the smoker and non-smoker group, why don't you just use a regression instead? Actually 14 is a very small sample size for whatever you do to generalize your comment on the population. That is may be another problem.
Nov
13
answered Likert Scale Analysis
Nov
11
comment Probability understanding 52 C 5
Welcome @Pedro.Alonso :)
Nov
11
comment Non significant Pearson correlations included in hierarchical regression?
You are most welcome @Simon
Nov
10
accepted Comparing the effect of a treatment that was optional for its receivers
Nov
10
comment Models and statistical methods for estimating populations?
I think for sampling with such sensitive issues, network sampling can be a good sampling method. Actually, there are many other methods. You may go through the sampling techniques and decide whichever may suit.
Nov
10
answered How to get the median of a row in SPSS?
Nov
10
answered Non significant Pearson correlations included in hierarchical regression?
Nov
10
comment Probability understanding 52 C 5
I think you are having confusion in only one part. If we pick two fruits from apple and orange, then there can be actually 4 combinations. You might wonder that 'Apple+Orange' and 'Orange+Apple' are the same things in 'combination-permutation' point of view. But if I assigned two numbers (say, randomly '2' and '5') along with selecting these combinations then are they the same? No they aren't. Then 'Apple+Orange' is actually 'Apple(2)+Orange(5)' and 'Orange+Apple' is actually 'Orange(2)+Apple(5)'. That's why 2^2 combinations are valid. Similarly 4^5 combinations for this problem are valid.
Nov
9
revised Probability understanding 52 C 5
added 340 characters in body
Nov
9
answered Probability understanding 52 C 5
Nov
8
comment Comparing the effect of a treatment that was optional for its receivers
Actually I don't have information about how many times they visited the advisers but I have the number of hours they spent with the advisers. In 2012 we did not have advisers but we collected data on demographic variables of the students. In 2013 we introduced advisers and also collected information about the demographic variables. Now based on the similar demographic variables in the 2013 data, we want to make a comparison with the 2012 data on retention of the students. We also want to see if introducing advisers in 2013 had any effect on retention.
Nov
8
comment Comparing the effect of a treatment that was optional for its receivers
Thank you for idea. It was really helpful. :)
Nov
8
answered Time series as cross-sectional data
Nov
7
comment Comparing the effect of a treatment that was optional for its receivers
They now simply want an answer along the lines of the following: "We weighted the 2013 cohort so that it was the same on key demographics as the 2012 cohort. In 2012, 42 (say) more students failed to return for the semester 2 2012 (attritition) than was the case with the matching dataset in 2013, presumably due to the interventions introduced in 2013." Is it possible?? Please help me with your kind suggestion (possibly in a different answer thread or by updating your existing answer). Thank you once again for your help.
Nov
7
comment Comparing the effect of a treatment that was optional for its receivers
Thank you and +1 for both of you. As per your suggestion we do not really need the 2012 data. 2013 data can alone provide us with propensity scores. But my bosses still think we need to weight the 2012 data so that it matches the 2013 data on some of the variables (8 variables in total) that may have been sensitive to the removal of the cap on student entry (or weight the 2013 data so that it matches the 2012 data). Then we can compare two different cohorts as if they were a proper control group for each other.