First draft of wave 4 changelog

Author: capnrefsmmat

docs/symptom-survey/coding.md

@@ -39,6 +39,9 @@ describe each survey response:
 * `weight`: The survey weight calculated by Facebook, for demographically
   adjusting estimates. See the [weights documentation](weights.md) for details
   on how to use these weights.
+* `wave`: Integer specifying the survey wave this respondent completed; see
+  below for the full list.
+* `UserLanguage`: Language the respondent completed the survey in.
 
 Coding details for each survey wave follow.
 
@@ -124,3 +127,107 @@ Files:
 * Consent question now requires the respondent be located in the U.S.
 * Item B2 now includes eye pain as a symptom.
 * Item C5 moved to be asked before item C4.
+
+## Wave 4
+
+Wave 4 will be deployed in late August. It will be available in English, as well
+as
+
+* *[translations to be completed]*
+
+Draft files [final versions to be posted when available]:
+
+* TODO (PDF)
+* TODO (Word)
+
+### Summary of Changes
+
+Wave 4 is a **major change** to the survey instrument. Some items have been
+removed and several new items have been added. Please review the changes
+carefully when you use responses from both waves.
+
+#### Consent Text
+
+The survey consent text has been altered to more clearly indicate that
+
+> Even if you are healthy, your responses may contribute to a better public
+> health understanding of the spread of the coronavirus pandemic and its effects
+> on public health and well-being. This may help improve our local and national
+> responses to the pandemic and our understanding of how it has affected
+> society.
+
+This is broader than the prior text, which suggested the survey was only to
+understand "where the coronavirus pandemic is moving", and ensures we obtain
+consent for the full range of uses for the survey.
+
+#### New Items
+
+* Following item B2, item B2c now asks which reported symptoms are "new or
+  unusual" for the respondent. This is intended to distinguish symptoms of
+  chronic conditions (such as allergies) from symptoms that may be COVID-related
+  (or due to influenza or another condition).
+* Item B7 now asks whether the respondent has sought medical care, such as
+  visiting a doctor or urgent care clinic. This will allow researchers to
+  understand how people seek medical care, and what fraction of people do so at
+  any given time.
+* Items B8, B10, B10a, B10b, B11, B12, and B12a more comprehensively ask about
+  testing, including whether the respondent has *tried* to be tested but could
+  not get a test, and why they were tested.
+* Item C13 asks about specific activities, such as working or going to a
+  restaurant, that the respondent may have done in the past 24 hours. Item C13a
+  asks whether the respondent wore a mask while doing them.
+* Item C14 asks "In the past 5 days, how often did you wear a mask when in
+  public?"
+* Item D6 now asks if the respondent is of Hispanic, Latino, or Spanish origin.
+  Note that this item **may not be available** in individual response files
+  until procedures to prevent reidentification of respondents are finalized.
+* Item D7 now asks the respondent's race. Note that this item **may not be
+  available** in individual response files until procedures to prevent
+  reidentification of respondents are finalized.
+* Item D8 asks for the highest level of school the respondent has completed.
+* Item D9 asks if the respondent has worked for pay in the past 4 weeks, while
+  items Q64 - Q80 ask the respondent to identify their occupation in a form
+  based on the [Standard Occupational Classification
+  System](https://en.wikipedia.org/wiki/Standard_Occupational_Classification_System).
+  (Each respondent will only see 2 of these items; the first requests a broad
+  category, the second a more detailed classification.)
+* Item D10 asks if any of the respondent's work for pay in the past 4 weeks was
+  outside their home.
+
+#### Changed Items
+
+* Item B2 now includes eye pain and chills as symptoms.
+* Item B2b now asks how long the respondent has had one *unusual* symptom, if
+  they report having one in item B2c. It no longer asks how long the respondent
+  has had *any* reported symptom.
+* Item C1 now separates type 1 and type 2 diabetes, coded separately. The coding
+  for both is different from the original coding for "diabetes", so they should
+  be treated separately.
+* Item C8 now asks how often the respondent has "felt isolated from others",
+  along with the prior items about anxiety and depression.
+
+#### Removed Items
+
+* Item A2b ("How many people are there in your household in total (including
+  yourself)?") has been removed, as have items D3, D4, and D5. They have been
+  replaced with item A5, which asks for the people in the household broken down
+  by age, *including* the respondent.
+* Items B3 and Q40, asking the temperature of respondents with fevers, have been
+  removed.
+* Item B4, asking if the respondent's cough involved mucus, has been removed.
+* Item B5 ("Have you been tested for COVID-19 (coronavirus) for your current
+  illness?") has been removed. It has been replaced with items B8-B12,
+  described above.
+* Item B6 is now subsumed by item B7.
+* Item C2 ("Have you had a flu shot in the last 12 months?") was not being used
+  by research partners and has been removed.
+* Item C3 ("In the past 5 days, have you gone to work outside of your home?")
+  has been replaced by a combination of items C13 and D9.
+* Item C4 is replaced by items Q64, Q68, and Q69.
+* Item C5 is replaced by items Q64, Q68, and Q69.
+* Item C7 ("To what extent are you intentionally avoiding contact with other
+  people?") is removed, replaced by the more specific items C13, C13a, and C14.
+* Item Q36 ("How much of a threat would you say the coronavirus outbreak is to
+  your household’s finances?") has been removed and replaced with item C15, for
+  consistency with the international version of the survey.
+* Item D1b ("Are you currently pregnant?") has been removed.