Archive for October, 2011

We found inspiration in the museums, canals, conference sessions and coffee shops. Perhaps a tremendous opportunity was missed because had we held the session in the coffee shops we might have achieved a new and vibrant interpretation of our beloved discipline.  Oh well, next time, Atlanta.

We are thrilled to have the next ESOMAR in that beautiful southern bell Atlanta. But we cannot help but think that there should be a special gathering of everyone who wished to attend Amsterdam and missed it or just those who plain old miss Amsterdam. We suggest a special cigar party—Kind of a spinoff from the CASRO party held at the annual each year. You figure it out.

We love Kristen Luck.  But Kristen, you out did yourself this time.  Now if Kristen wasn’t smarter than every other banana in the bunch we would assume that she meant what she said and said what she meant.   However, that’s what she said.   We’ll try to leave out any talk of lofty statistics in your presence Kristen.

Panel Study,  L. Ren, L. Li,     Q. Xu, J. Yan, Z. Qiu,  Peking University China.

Proper occupation assignment is often critical in interpreting social science and market research data.   It is a dilemma that haunts researchers around the world, one not particularly limited to China.   The challenges inherent in different modes of data entry, complicated by differences in the nuances of meaning that can be found in fine grain interpretation of occupations are a nightmare.

Here two methods of entering/interpreting the data, drawn from a Computer-Aided Personal Interviewing (CAPI) survey, were employed:  entering the data live by the interviewer and entering the data after the fact by experienced coders.   The results differed between the two methods significantly and meaningfully.   Imagine how different such results must be between respondent entered data in an online survey and interviewer entered data in a telephone interview.   It’s a frightening prospect.

Academicians are very sensitive to the sample frame.   Market researchers should be as well but we have adopted the non-probabilistic access panels as our main online sample source.   It was refreshing to see how many talks were devoted to probabilistically created panels.   For us it was a sampling heaven.

Based on the talks we heard, the Europeans are still pioneering on their use of probabilistic panels.  No large commercial panels of the type have made their appearance and those that exist are academic or government based.   They are still testing the waters but are clearly sensitive to the issues.

A battery of studies came out of work performed by the CentERdata Tilburg University, the Netherlands.   In 2007 CentERdata started the LISS panel (Longitudinal Internet Studies for the Social Sciences), which is “an online panel which is based on a true probability sample of households.”  It is representative of the Dutch population and consists of 5000 members.  This may be small by commercial standards but apparently they are producing quite a bit of research as we were privileged to listen to three good presentations.

The LISS panel began with a random sample of some 8000 respondents in 2007.  Those who did not have internet connection were provided with a computer and a connection.    By 2010 approximately 5000 were still active (see www.lissdata.nl).  The “attriters” were found to be different from the “stayers” in that the latter were more conscientious, open and female.   Those that tend to leave the panel were found to be Christian Democrats less often.

“Lurkers”, those who come and go in terms of completing some or many of the 25 monthly waves of the subject study, tended to have a lower level of education.

Those who abandoned the panel entirely, as opposed to the lurkers who came and went, were labeled “monotonic attriters”.

This is the first paper that we know of, besides our own, that drew a behavioral distinction between attriters and stayers.   We consider this to be an important finding;   One that we do not feel that we can do justice to here.   But the concept of a changing panel population created by differential attrition is one that we must begin to address.

The big conferences have a pattern of their own.   Soon you find your groove and recognize those who share the same interest that you do.  Time and again they take seats in the same crowded room or are found waiting outside of the same room where the talk has gone over time.

Often the colleagues that we make are fellow speakers who are advancing their craft.   Like us they are there to soak up content.   It is with them that the side discussions can become lively.   You exchange business cards or ideas.   Whatever it is, like trout going upstream, you find yourselves climbing the same staircases.

The academic buildings are rather drab.   We are not speaking of fancy conference centers here.   There are no bars to gather in, no plush carpets, or chandeliers.   Instead, un-air-conditioned, formed concrete towers with lecture halls instead of banquet halls.   Don’t fool yourselves, while the trappings may not be fancy, the intellectual stuff really is.

Oh, don’t get me wrong, some of the talks can be deadly.  One professor droned on in a monotone voice heavily laden with equations.  A doctoral student sat listlessly reading from her notes.   One government type got up there and winged it without slides but handed out a photocopied list of variables for every other person.   When you get pinned into a room by a crowd and someone comes unprepared or presents poorly it is frustrating.    Most of the talks are vibrant and stimulating.   Some of the graduate students are so wired into what they do that they speak their subject matter with a natural fluency that can only be admired.