In SQL I have a column called "answer", and the value can either be 1 or 2. I need to generate an SQL query which counts the number of 1's and 2's for each month. I have the following query, but it does not work:
SELECT MONTH(`date`), YEAR(`date`),COUNT(`answer`=1) as yes,
COUNT(`answer`=2) as nope,` COUNT(*) as total
FROM results
GROUP BY YEAR(`date`), MONTH(`date`)
I would group by the year, month, and in addition the answer itself. This will result in two lines per month: one counting the appearances for answer 1, and another for answer 2 (it's also generic for additional answer values)
SELECT MONTH(`date`), YEAR(`date`), answer, COUNT(*)
FROM results
GROUP BY YEAR(`date`), MONTH(`date`), answer
Try the SUM-CASE trick:
SELECT
MONTH(`date`),
YEAR(`date`),
SUM(case when `answer` = 1 then 1 else 0 end) as yes,
SUM(case when `answer` = 2 then 1 else 0 end) as nope,
COUNT(*) as total
FROM results
GROUP BY YEAR(`date`), MONTH(`date`)
answer=1 then answerelse null end). It is more useful when you have joins that will make the same rows appear more than once and you have a distinct value in answer (using sum(1) will add it several times for the same master row). Not useful in this case, but worth mentioning :)SELECT year,
month,
answer
COUNT(answer) AS quantity
FROM results
GROUP BY year, month, quantity
year|month|answer|quantity 2001| 1| 1| 2 2001| 1| 2| 1 2004| 1| 1| 2 2004| 1| 2| 2
SELECT * FROM results;
year|month|answer 2001| 1| 1 2001| 1| 1 2001| 1| 2 2004| 1| 1 2004| 1| 1 2004| 1| 2 2004| 1| 2
