[acs-r] Warning message

[Ezra Haber Glenn]

There are sometimes when data is missing from the ACS for a particular
geography and table -- I think when there were too few (or no)
observations to report.  When this happens, the census API returns a
special code, which the acs.R package converts to NA (a reasonable
value for a missing values.)  

The warning is simply R telling you that there you've applied some
operations to NAs, which is something you might want a warning about.
(When these missing values get manipulated -- e.g., when the script
converts the NA margin of error into a standard deviation -- a
division process -- you get NAs propagating.  This is perfectly
appropriate, but the warning is helpful in R for other things, where
you might be surprised -- e.g., 3+5+NA=NA, not 8.)  

If you look at all the data you fetched (and hot just the "head"),
you'll see a number of times this occurred -- actually, quite a few:

> sum(is.na(estimate(acs.data)))
[1] 1664
> sum(is.na(standard.error(acs.data)))
[1] 1671
> 

Taking one example -- ZCTA 60932 from your acs.data set -- you'll note
that this is not simply an acs.R or Census API issue -- if you check
FactFinder for this table and ZCTA, you will see that the fields are
blank:

<http://factfinder2.census.gov/bkmk/table/1.0/en/ACS/12_5YR/B01002/8600000US60932>

As the notes there explain:

   Explanation of Symbols:

    An '**' entry in the margin of error column indicates that either
    no sample observations or too few sample observations were
    available to compute a standard error and thus the margin of
    error. A statistical test is not appropriate.

    An '-' entry in the estimate column indicates that either no
    sample observations or too few sample observations were available
    to compute an estimate, or a ratio of medians cannot be calculated
    because one or both of the median estimates falls in the lowest
    interval or upper interval of an open-ended distribution.

(The "**" and "-" values are converted to NAs by acs.R.)

My guess is that in your case you don't actually care that much about
these NAs-- they are probably empty or tiny zip codes -- but if you
next started to combine your ZCTAs, you'd actually start losing data
-- the package doesn't treat NAs the same as 0s, for good reasons.
Depending on what you're doing with this data, you might want to first
investigate the NAs and decide what to do with them -- remove them,
convert them to zero, etc.

Hope that helps.

--Ezra


At Sat, 1 Mar 2014 13:38:06 -0800, arilamstein wrote:
> 
> When I try to get median age of all ZCTAs I get a warning message that I don't understand:
> 
> acs.data = acs.fetch(geography=geo.make(zip.code = "*"), 
>                      table.number = "B01002", 
>                      col.names = "pretty", 
>                      endyear = 2012, 
>                      span = 5)
> 
> Warning message:
> NAs introduced by coercion 
> 
> The resulting data.frame appears OK to me:
> 
> > head(acs.data)
> ACS DATA: 
>  2008 -- 2012 ;
>   Estimates w/90% confidence intervals;
>   for different intervals, see confint()
>             Median Age by Sex: Median age -- Total: Median Age by Sex: Median age -- Male
> ZCTA5 00601 36.4 +/- 0.5                            34.7 +/- 0.9                         
> ZCTA5 00602 38 +/- 0.4                              37.2 +/- 0.5                         
> ZCTA5 00603 38.4 +/- 0.6                            36.7 +/- 0.7                         
> ZCTA5 00606 36.8 +/- 2                              36.8 +/- 3.2                         
> ZCTA5 00610 38.6 +/- 0.6                            36.1 +/- 0.8                         
> ZCTA5 00612 38.1 +/- 0.6                            36.9 +/- 1.2                         
>             Median Age by Sex: Median age -- Female
> ZCTA5 00601 37.7 +/- 1.5                           
> ZCTA5 00602 38.7 +/- 0.6                           
> ZCTA5 00603 40.1 +/- 0.6                           
> ZCTA5 00606 36.8 +/- 2.9                           
> ZCTA5 00610 40.5 +/- 0.5                           
> ZCTA5 00612 39.2 +/- 1     
> 
> Can someone tell me what this warning means, and whether I need to be concerned about the
> resulting data?
> 
> Thanks.
> 
> Ari
> 
> 

--
Ezra Haber Glenn, AICP
Department of Urban Studies and Planning
Massachusetts Institute of Technology
77 Massachusetts Ave., Room 7-337
Cambridge, MA 02139
eglenn at mit.edu 
http://dusp.mit.edu/faculty/ezra-glenn | http://eglenn.scripts.mit.edu/citystate/
617.253.2024 (w)
617.721.7131 (c)