average income (Louisiana data)

Our research group has recently moved up to V3 and when we started working on one of our first models, we found some of the summary results confusing. Specifically, we noted that St. Bernard Parish has a very high per capita income of more than $195,000 according to Implan when the Census puts per capita income for the parish at $16,718 in 1999 and median income of $39,106 in 2008. Of course, 1999 is a long time ago and I know the differences between per capita and median so we can't directly compare 2008 median income with the number in Implan. But, the disparity is just too large. St. Bernard is a low income parish and there's no way that per capita income could be so high. Is there any explanation? Is there something I've done wrong with my settings? Is there something unusual about how per captia income is derived within Implan that might lead to this? We have also found similarly skewed data for other parishes and are concerned about the integrity of the data.
Was this post helpful?
0 out of 0 found this helpful

Comments

1 comment

  • The difference you are referring to is the difference between Average Personal Income and Median Money Income (Census). For the difference see this [url=http://implan.com/v3/index.php?option=com_docman&task=doc_download&gid=134&Itemid=138]link[/url]. Personal Income comes from REIS and number of households come from the deciennial Census and is adjusted for each year of data based on the change in the number of household units from 2000. Looking at a map, I see that St Bernards is a Mississippi river delta county at the border of Louisiana and Missippi. I wonder if Katrina wiped out enough unused household units to skew the households count? In this case I would also suspect the Average Personal Income per household reported. This will not affect the databases as long as the distribution of the household in the various household classes do not change (eg, 10% of the households are in the highest household class). If the distribution changes the gross purchases by each of the household classes would change - but the overall household spending would still be controlled to the total REIS number.
    0

Please sign in to leave a comment.