Survey Automation Report and Workshop Proceedings (2003) / Chapter Skim
Currently Skimming:
Understanding the Documentation Problem for Complex Census Bureau Computer Assisted Questionnaires
Understanding the Documentation Problem for Complex Census Bureau Computer Assisted Questionnaires
Pages 83-96
The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.
From page 83... ...
Piazza's Web site from which he delivered this talk, and which contains links to additional information on both IDOC and TADEQ is still accessible at http://sda.berkeley.edu/present/automation. From that page, links are available to the specific pages referenced in these remarks; we have excerpted some of those in this printed report.
|
From page 84... ...
XML representation of the Blaise instrument. So they're thinking along these lines also, to some extent.
|
From page 85... ...
i9GoTO refers to the command contained in some computer programming languages (the best known example of which is probably the BASIC language) which allows program execution to immediately jump to another line or another portion of the program.
|
From page 86... ...
That is a possibility. But I'm going to switch gears here now, and let's look at an individual item one of these health insurance items we were lolling at titem ME 163.
|
From page 87... ...
ENTER " N " FOR NO PAYMENTS Q3 dollars Description of Instrument Flow Universe: all adults How to Get to This Item The preceding decision point: ~14 Default preceding item: M[E1L5 Where You Can Go From This Item Based on the value of this item: Go to ME18 if this item = N Go to ME18 if this item = 1-99999 Otherwise go to the next item: ME17 87 Figure 11-5 Item ME16 from Instrument Document (IDOC) for Wave 6 of the Survey of Income and Program Participation (SIPP)
|
From page 88... ...
In other words, you start with a paper questionnaire, and on the paper questionnaire, you have something where if you answer "yes" there's a little arrow that says "go here" or another arrow that says "no, go there." And what CASES did in the development 24Specifically, if the respondent answered "don't know" or "refuse" to giving a specific amount paid over the preceding year, they are routed to a question that asks for a (presumably easier to estimate) categorical value, e.g., "less than $500," "$500 to $1000," and so on.
|
From page 89... ...
Now this is not obvious from the instrument itself. Blat the program that plats together this whole set of HTML files goes through all of the destinations in the instrument.
|
From page 90... ...
Now it would be possible and there are hooks for the author to put a lot of this information in as you go. And ideally that's the time to 25 The "more detail" screen for ME14 includes a complicated sequence of set-up commands that would insert appropriate phrases (e.g., regarding diabetic equipment)
|
From page 91... ...
the authors don't ever have to use the data there's very little incentive to malce things easier for people down the pilce when they're under immediate pressure to get their own job done. So part of it's an organizational thing.
|
From page 92... ...
, who's going to go get the SIPP file 26IASSIST is the International Association of Social Science Information Service and Technology, and its membership includes social science researchers, information specialists, and methodologits/computing specialists.
|
From page 93... ...
PIAZZA: There's an additional problem, too, and that is especially in terms of federal data sets all of these variables are derived, or censored, or recoded, or something, so that the data set you get is far removed from the original questionnaire. SO, one thing that we are working on in terms of the documentation of the data sets is to allow links back to the document for the questionnaire, which is related but not the same thing.
|
From page 94... ...
I haven't heard anybody say anything about this today but there's an effort at the Census Bureau to put together a corporate metadata repository, which would be a place where you could put those specifications. And it drives part of the software development process be it CASES or Blaise to get the labels onto data at that point of the process, to malce it easier to either analyze data or to document it on the Web .
|
From page 95... ...
ESO, ideally, this migration to a common intermediate format might malce it possible to develop more universally applicable documentation tools.] And maybe more would get done because, remember, this is a very small market.
|
From page 96... ...
Which, if you've got these in some nice coordinated software package .
|
Key Terms
This material may be derived from roughly machine-read images, and so is provided only to facilitate research.
More
information on Chapter Skim is available.