Skip to content

Commit a5bca3c

Browse files
authored
Merge pull request #901 from cmu-delphi/survey/better-docs
Improvements to CTIS documentation
2 parents eaf8d01 + bcbc55b commit a5bca3c

File tree

4 files changed

+465
-352
lines changed

4 files changed

+465
-352
lines changed

docs/symptom-survey/coding.md

Lines changed: 23 additions & 15 deletions
Original file line numberDiff line numberDiff line change
@@ -7,10 +7,17 @@ nav_order: 6
77
# Questions and Coding
88
{: .no_toc}
99

10-
The COVID-19 Trends and Impacts Survey (CTIS) has been deployed in several waves.
11-
We have tried to ensure the coding of waves is consistent. This page provides
12-
the full survey text and coding schemes.
13-
10+
The COVID-19 Trends and Impacts Survey (CTIS) has been deployed in several
11+
waves. We have tried to ensure the coding of waves is consistent. This page
12+
provides the full survey text, coding schemes, and history of survey waves and
13+
revisions.
14+
15+
<div style="background-color:#f5f6fa; padding: 10px 30px;"><strong>Comprehensive
16+
codebook:</strong> Our <a href="codebook.csv">codebook (CSV)</a> lists all
17+
questions and answer choices across all waves of the survey. See below for
18+
details on the <a href="#basic-coding-rules">basic coding rules</a> and the <a
19+
href="#comprehensive-codebook">formatting of the codebook</a>, and narrative
20+
explanations of the reasons for each instrument change.</div>
1421

1522
## Table of contents
1623
{: .no_toc .text-delta}
@@ -45,12 +52,13 @@ The following columns were added beginning when Wave 4 was deployed:
4552
* `wave`: Integer specifying the survey wave this respondent completed; see
4653
below for the full list.
4754
* `UserLanguage`: Language the respondent completed the survey in.
48-
- `fips`: The *primary* county FIPS code corresponding to the ZIP code selected
49-
by the respondent. Note that ZIP codes can cross county and even state
50-
boundaries; if a respondent's ZIP is in multiple counties, the FIPS reported
51-
in this column corresponds to the county the ZIP overlaps most with. If a ZIP
52-
is not more than 50% in any one county, or if their reported ZIP code cannot
53-
be found, `NA` is reported.
55+
- `fips`: The *primary* county [FIPS
56+
code](https://en.wikipedia.org/wiki/FIPS_county_code) corresponding to the ZIP
57+
code selected by the respondent. Note that ZIP codes can cross county and even
58+
state boundaries; if a respondent's ZIP is in multiple counties, the FIPS
59+
reported in this column corresponds to the county the ZIP overlaps most with.
60+
If a ZIP is not more than 50% in any one county, or if their reported ZIP code
61+
cannot be found, `NA` is reported.
5462

5563
Beginning in Wave 11, the `module` column indicates which module the respondent
5664
was randomly assigned to. See the [survey modules and randomization](modules.md)
@@ -93,11 +101,11 @@ following rules:
93101

94102
## Comprehensive Codebook
95103

96-
The [codebook](codebook.csv) describes availability of metadata and question
97-
fields by survey version. A row is included for each field available in the
98-
microdata files for a given survey wave. We recommend using the codebook as a
99-
reference while working with the survey microdata, to ensure you interpret each
100-
question and answer choice accurately even as items were revised between
104+
The [codebook (CSV)](codebook.csv) describes availability of metadata and
105+
question fields by survey version. A row is included for each field available in
106+
the microdata files for a given survey wave. We recommend using the codebook as
107+
a reference while working with the survey microdata, to ensure you interpret
108+
each question and answer choice accurately even as items were revised between
101109
survey waves.
102110

103111
If a field is available for multiple waves, it is listed separately for each

0 commit comments

Comments
 (0)