Monday, November 2, 2009

Monday morning announcements

{Made updates below:}

Hi all,

* We are doing ANOVA and ANCOVA in the second half of lecture 7. Lecture 8 will be on Factor and Cluster analyses. Lec 9 will be on Logit and Discriminant and other miscellaneous MKTG models. Shall see if conjoint can be worked in somewhere as well.

Point is, ALL these methods can be traced back and expressed as modeling problems that we explored in the Modeling primer.

Understanding a basic Y = f(X) +e kind of framework goes a long way in formulating, organizing, and operationalizing the data analysis problem.

Pls revise the modeling primer portion of lec 5 as a warm up for the Quant part of the course. Again, for non-technical people, its OK to not go into details of the Primer. One can focus merely on using SPSS well, interpreting its results and moving on. That is all that will be expected for the exam.

* Pls ensure SPSS trial version is up and running for lecture 7 and subsequent lectures. The software comes in the nick of time from the data analysis lectures and the project point of view.

* Have received suggestions that the data collection deadline be moved to Friday midnight so that data analysis can start Saturday noon onwards itself. I tend to agree, just that my concern is will I get enough of a sample base by Friday midnight to close the survey and proceed from there?

* Shall announce a lucky draw of Rs 3000/- gift cheque for respondents who successfully stuck through and completed the survey. So pls mention this incentive when contacting potential respondents from within your networks. If this incentive scheme helps get a few score more complete responses, it would well be worth it. Also, it would help in getting more personal details from the respondent like name and email ID.

Further, if any respondent asks, pls feel free to say that result summaries of the survey can be shared with them should they be interested. This itself may act as an incentive of sorts for auto enthusiasts.

* Shall talk to the IT dept to see if IP addresses from within ISB can be IDed per se. Am no IT guy and don't quite know if this can be done. I would rather janta e-survey non-ISB folk for this one.

* Sorry I lost my cool of sorts yesterday. Was working away the whole day into the late evening on the survey and to then have some smart alec quote my mail to me to say 'I'm spamming his mailbox' and that 'MKTR has this problem in general' I guess just about tripped me over. I don't know who wrote that, I don't want to know either. That incident is done with and behind me now. However the anonymous commenting restrictions it has spurred will continue.

Sudhir

No comments:

Post a Comment

Constructive feedback appreciated. Please try to be civil, as far as feasible. Thanks.