[brianchu]

I have the opposite assumption: the raw data is usually more reliable than an editorial. And I confirmed it: the raw table is more reliable since it is the same survey across different years. The article is inconsistently comparing numbers from 2 different surveys. The 2018 figure (2.63) is from the "American Community Survey" and the 2010 figure (2.58) is from "Census SF1 data".

But the 2010 "American Community Survey" says the average houshold size is 2.63 (https://data.census.gov/cedsci/table?q=b25010&tid=ACSDT1Y201...), so for this survey the trend is flat.

[dv_dt]

The pew article is more comprehensive than those two surveys. They also point out supporting data from

https://www.census.gov/library/publications/2019/demo/p60-26...

From which they conclude "In 2019, 20% of households are shared households, up from 17% in 2007."

https://www.pewresearch.org/fact-tank/2018/04/05/a-record-64...

from which they mention "By 2016, 20% of Americans lived in a multigenerational household, up from 12% in 1980"

One survey is one survey, multiple data points from a set of different perspectives is more comprehensive an analysis and less susceptible to single measurement methodology errors.

The main thing I take from this is that movement of a household income number is fairy meaningless unless you also know what household size number is measured or calculated with the income number.