SQL unknown columns? - sql

Question

Im new to SQL and am having difficulty with this query:
The average elevation of each Texas county which contains one or more zip code with a population over 20,000.
This is what I have so far:
SELECT county, zip_code, state, AVG(elevation)
FROM zip_codes
WHERE state=’TX’
GROUP BY county
HAVING population > 20000
ORDER BY county;
ERROR 1054 (42S22): Unknown column '’TX’' in 'where clause'
But when I select state from the zip_codes database it is there.

Answer 1

You're using wrong single quotes. Should be
WHERE state = 'tx'
instead of
WHERE state = ’tx’
Yours are ... where from? Some editor like MS Word or similar, which creates those "fancy" quotes. Well, don't use them.
Apart from that, columns that aren't aggregated should all be part of the GROUP BY clause:
group by county, zip_code, state

Answer 2

Fixing the quote issues is a must-have, as answered by Littlefoot.
However, your query is still invalid Oracle SQL, since:
the select clause has two non-aggregated columns that are not part of the group by clause (zip_code and state)
same goes for column population in the having clause, that is neither aggregated nor part of the group by clause
For this assignment:
The average elevation of each Texas county which contains one or more zip code with a population over 20,000.
Assuming that each record in the table corresponds to to a different zip_code (which seems relevant, given that the table itself is called zip_codes), you could phrase:
select county, state, avg(elevation) avg_elevation
from zip_codes
where state= 'TX'
group by county, state
having max(population) > 20000
order by county;

Answer 3

SELECT county, zip_code, state, AVG(elevation)
FROM zip_codes
WHERE state='TX'
GROUP BY county, zip_code, state
where population > 20000
ORDER BY county;

Answer 4

Without a data sample it's difficult to understand precisely what's going on, but I'll take a stab at it.
Your assignment says you are to compute "The average elevation of each Texas county which contains one or more zip code with a population over 20,000". Let's break down the elements of the assignment you need to find:
Average elevation
Of each county in Texas
Which has one or more zip code
With population > 20,000
And we've got a table of zip codes, with county, state, and elevation.
So, OK - first problem: find all the zip codes in Texas:
SELECT *
FROM ZIP_CODES
WHERE STATE = 'TX'
OK, easy so far. Now, limit it to zip codes with a population over 20,000:
SELECT *
FROM ZIP_CODES
WHERE STATE = 'TX' AND
POPULATION > 20000
Cool - but this gives us an entry for each zip code in the county with a population > 20000. All we care about are the individual counties which have AT LEAST one zip code with a population over 20K. So how about
SELECT DISTINCT COUNTY
FROM ZIP_CODES
WHERE STATE = 'TX' AND
POPULATION > 20000
Great. Now we need to compute the average elevation in each county returned by the above query:
SELECT COUNTY, AVG(ELEVATION)
FROM ZIP_CODES
WHERE STATE = 'TX' AND
COUNTY IN (SELECT DISTINCT COUNTY
FROM ZIP_CODES
WHERE STATE = 'TX' AND
POPULATION > 20000)
GROUP BY COUNTY
ORDER BY COUNTY
And there's your answer.