Edit post

Edit your reply below.

Post Body
Enter your post below.
[quote user="schlimm"] [quote user="Sam1"]None of the polls that you references, if I remember correctly, used statistical sampling. Moreover, even if they did, all of them relied on telephone interviews, the results are suspect.  The consultants that we used (McKinsey & Company, Booz, Allen & Hamilton, Planmetrics, Anon Hewitt) would not rely just on the results of telephone interviews.[/quote] I would politely suggest you re-examine the last page of this PP. <a href="http://www.texasrailadvocates.org/pg_utstudy.asp">http://www.texasrailadvocates.org/pg_utstudy.asp </a>  You will see that, contrary to your recollection, "statistical sampling" was of course used.   University polling groups such as those at Univ. of Chicago or Univ. of Michigan, to name two, are as good or better than polling done by (or for) consulting companies. The use of telephone and now internet-based surveys is widely used  with decent validity as long as basic principles are adhered to: randomization, avoidance of response bias, making sure sample is representative, and avoidance of self-selected samples. [/quote] I stand partially corrected. The DFM polls used a valid statistical sampling technique.  The Texas Poll, which is the one that I focused on, took a random sample of 2000 Texas residents. The researches have not disclosed the level of confidence they used and how they selected the size of the sample. TRA) - Austin - A wide-ranging study conducted by a research group at the University of Texas shows that an overwhelming majority of Texans surveyed.  This is the introduction to the Texas study.  If this is not restricting the results to those surveyed, I don't know why the researchers would have included it in the introductory paragraph of their presentation.  Unless it was written by someone else, and they missed it. The size of a sample is determined by the confidence level, tolerable or expected error rate, etc.  It is also impacted by any required replacements.  It appears that the UT researches chose 2,000 arbitrarily. And used the results of the 34 per who responded to draw their conclusions.  As they say in the introduction to their presentation, they are presenting the results of those surveyed, which is different than saying that they are confident that the population of Texas says such and such. How would you know about the qualifications of the consulting companies that I referred to?  How would you know whether they are as good as or not as good as university researchers? All the studies contain a weakness in my view.  The researchers did not conduct face to face follow-ups nor did they engage any focus groups.  That is a serious shortcoming. The biggest shortcoming lies in the fact that irrespective of what people say, at the end of the day very few of them travel intercity train or use public transit.  And although some say that they would in the future, how many actually do is problematic. Equally interesting, a significant majority of the Texas respondents said that they would be willing to spend additional tax dollars on better transport options.  Did the researchers tell them how much it would mean out of their pockets?  Doing so frequently generates a different response. I suspect that many of the respondents said that they favor the current level of train service or would like to see it increased because it has very little impact on their pocketbook. This is true for Amtrak as a whole. Although it required a cash subsidy of $1,620,533,000 in FY13, it only amounted to $7.28 for every person over 18 or $17.46 for every 2011 federal income filer with a tax obligation . From my point of view the Texas poll is low value. I regret engaging in this discussion.
Tags (Optional)
Tags are keywords that get attached to your post. They are used to categorize your submission and make it easier to search for. To add tags to your post type a tag into the box below and click the "Add Tag" button.
Add Tag