ACS Data Users Group

I've looked at a few Public Health sites and I couldn't match their number to S1101, then I realized I couldn't match S1101 measure to anything even other numbers in S1101. I must be doing something wrong.

https://data.census.gov/table?q=S1101&g=0100000US&y=2020&d=ACS+5-Year+Estimates+Subject+Tables&tid=ACSST5Y2020.S1101&moe=false

It shows 41.8% households married have a child under 18. I think that's 41,8% of all households? Not 41.8% of households with a child under 18, but I'm not sure. In any event how can you have male at 55 and female at 62, there's no way the math will work? However, either way, it should be kids in married household = a kids in male households b, female households c , non-family =d. a+b+c+d = all households with kids under 18.

Tom

The way to read this is first look at the top column header to know what universe you're starting with. If the column is "Married couple family households", then everything in that column will only include married couple family households (eg. the row that says "Total households", for that column, is total households THAT ARE married couple family households.

Then, when looking at a row with a percentage, check what's above the indented section to see what the denominator for that percentage is. So "Households for with one or more people under 18 years" is under "Total households"... so if it's in the column for "Married couple family households", then that means, of all the households that are married couple households, what percentage have people under 18 years?

I think you might be misunderstanding with your contention that the math doesn't work. The percentages of male- and female-headed no-spouse family households with kids are independent of each other, and both are independent of the percentage of married family households with kids.

Your assumption that a+b+c+d=all works for the absolute numbers (count up the total number of households and they equal the total in the first column). It doesn't work for percentages, since they need to be weighted by the size of the population. 41.8% of married family household have kids, 55% of male-no-spouse family households have kids, 62% of female-no-spouse family households have kids. Together, that's about 79 million households. But then, there are 42 million non-family households, of which less than 1% have kids. So that brings down the total percentage for all households to 30.7%.

Hi, again appreciated all the help in this forum. I am still trying to make the math work but can we first agree'

Under 6 year only

Under 6 years and 6 to 17 years

6-17 years only

does not work in any version of math?

Under 6 years and 6 to 17 years = 0-17 and the percentage can't be less than 6-17 years old.

I'm not that smart but sometimes it's an advantage because I really need to work things out so I understand.

OK after much effort I think I understand

under 6 only = all children under 6

under 6 and 6-17 = some children under 6 some older

6-17 only = no children under 6

No, under six and six to 17 means at least one child under six plus at least one child 6 to 17. It does not mean one child 0 to 17.

"how can you have male at 55 and female at 62, there's no way the math will work"

What? Lots of marriages have one spouse 7 years older than the other.

Bernie is right. 41.8% of married couple households have a child 18 years or younger. 55% of male household, no spouse have a child 18 years or younger, and 62.2 percent of female household, no spouse have a child 18 years or younger. Why do you think these all need to add up to 100%? You are comparing different household types.

Yes, I understand now. I have not seen much mix and match in the tables and was thinking along the standard lines of most age groups I see here. I always try to do the match numbers and percent to make sure I understand, otherwise I could have the wring assumption which makes this a good example. Maybe I'll take my final measures list and pay someone in Upwork who knows the data to verify my data and assumptions.