Question

I have a table with 3 columns being country, user_id, gender.
The user_id just increments, there are a number of different countries and the gender is either male or female. Similar to this

user_id  country  gender
1        japan    male
2        peru     female
3        japan    female
4        fiji     male
5        peru     female

I want to run one query to return a unique list of countries, the total number of user_id‘s for the country and the male and female numbers for each country. so the results would look like:

country total male female
peru    2     NUL  2
japan   2     1    1
fiji    1     1    NUL

I have tried a number of differnet queries but can’t return all the results in one query and don’t want to use 2 queries if possible.

this is my last try – I know its a mess but I was trying anything:

SELECT DISTINCT a.country AS country, COUNT(b.gender) AS male, COUNT(c.gender) AS     female, COUNT(a.user_id) AS total
  FROM userattributes as a
  LEFT JOIN userattributes as b ON a.user_id=b.user_id
  LEFT JOIN userattributes as c ON a.user_id=c.user_id
  WHERE b.gender='male' AND c.gender='female'
  GROUP BY country;

Can anyone do the above in one SELECT?

thx

Answer 1

There’s no need to JOIN the table with itself. Using a CASE expression with the SUM aggregate would suffice.

SELECT  country
        , COUNT(*) AS total
        , SUM(CASE WHEN gender = 'male' THEN 1 ELSE 0 END) AS male
        , SUM(CASE WHEN gender = 'female' THEN 1 ELSE 0 END) AS female
FROM    userattributes
GROUP BY
        country