counting number of times a particular level and then aggregating it the number of times in new variable - sql

Question

Counting number of times a particular level (in transaction data) and then aggregating it the number of times in new variable (under one row per customer)
I have 2 levels to solicitation method, phone and email. I have created 2 new columns which count the number of times phone or email happened per id. Right now I have transaction data and cant figure out how to go about it. the data is on left, what I want is on right. I am okay with both kinds of output on right side.
So far I tried this. returns error
create table d.email as
select ID, email_count
from d.emai
where email_count = (select count (*)
from d.email
group by ID
having SolicitMethod = 'Email' );
quit;

Answer 1

I am not sure what you really want to do, but you can fix the syntax error by making the subquery a correlated subquery:
create table d.email as
select ID, email_count
from d.emai e
where email_count = (select count(*)
from d.email e2
where e2.SolicitMethod = 'Email' and e2.id = e.id
);
I assume the reference in the first from should be d.emai.

Answer 2

The first output can be obtain with this query:
It groups rows by id, and then count how many rows are on each SolicitMethod
SELECT id
, SUM(CASE
WHEN SolicitMethod = 'Email' THEN 1
ELSE 0
END) count_email
, SUM(CASE
WHEN SolicitMethod = 'phone' THEN 1
ELSE 0
END) count_phone
FROM d.email
GROUP BY id
This second output query depends of your dbms and availability of analytics function:
it count on each rows the count of sollicitMethod of each group of id
SELECT id
, SUM(CASE
WHEN SolicitMethod = 'Email' THEN 1
ELSE 0
END)
OVER (partition BY id) count_email
, SUM(CASE
WHEN SolicitMethod = 'phone' THEN 1
ELSE 0
END)
OVER (partition BY id) count_phone
FROM d.email