Ok, first off I'm starting this in the Access forum as I'd just assume be able to do it locally rather than having to go through our DBA to create something on the SQL Server to work with. However, given the nature of what I'm trying to achieve (which I'm not even sure is remotely possible at this point), it may end up there instead. Here's the basic rundown:
I'm in the process of conducting a survey for our company using one of the online sites (SurveyGizmo in this case) for the process. It's more of a 360 Rating than survey, but in effect, the participant is asked to select from a list of 130 people (minimum 8, but someone could pick all 130) to rate. They will then be asked the same 4 questions on each person they selected.
This part is fine, the trouble comes in building reports out. SurveyGizmo's built in reporting is very respondent-centric, where I need to generate more target specific data. The best I can get from them is a CSV style export. The challenge that their system exports 1 row per respondent, so the table has every person/question as the fields (130 people * 4 questions = 520 fields, plus the respondent data). What I ultimately need is a row per person rated with the respondent ID, the ratee ID, and the ratings on the 4 questions (6 fields).
So, if I have a table that looks something like this:
Could I use some manner of fun and exciting query to convert it something more like this:
I'm expecting somewhere around 1500+ responses, so I'm desperately hoping I can do this programatically rather than by hand.
Thanks in advance to anyone who has an idea how this could be achieved.
I'm in the process of conducting a survey for our company using one of the online sites (SurveyGizmo in this case) for the process. It's more of a 360 Rating than survey, but in effect, the participant is asked to select from a list of 130 people (minimum 8, but someone could pick all 130) to rate. They will then be asked the same 4 questions on each person they selected.
This part is fine, the trouble comes in building reports out. SurveyGizmo's built in reporting is very respondent-centric, where I need to generate more target specific data. The best I can get from them is a CSV style export. The challenge that their system exports 1 row per respondent, so the table has every person/question as the fields (130 people * 4 questions = 520 fields, plus the respondent data). What I ultimately need is a row per person rated with the respondent ID, the ratee ID, and the ratings on the 4 questions (6 fields).
So, if I have a table that looks something like this:
Code:
RespID EmpX-Q1 EmpX-Q2 EmpX-Q3 EmpX-Q4 EmpY-Q1 EmpY-Q2 EmpY-Q3 EmpY-Q4 etc...
1234 2 3 3 2 3 3 3 3
Could I use some manner of fun and exciting query to convert it something more like this:
Code:
RespID RateeID Q1 Q2 Q3 Q4
1234 X 2 3 3 2
1234 Y 3 3 3 3
I'm expecting somewhere around 1500+ responses, so I'm desperately hoping I can do this programatically rather than by hand.
Thanks in advance to anyone who has an idea how this could be achieved.