If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
Thread Tools | Display Modes |
#1
|
|||
|
|||
Likert Scale data- how to enter in Access
I have created a database to record responses to a questionnaire in which
respondents must answer YES, NO or I Don't know for each question. For EACH question, their response gets a different weight- for example, sometimes a YES response = 3, sometimes = 1. SO I have created an "answer code table" in which each question has a unique ID, and then I entered the correct values for YES, NO, and I DON"T KNOW responses. Then I created a "survey" table record responses, where each survey has a unique ID. SO now what I want to do, is link the answer key table with each survey response, so that I can easily enter the survey data and calculate each respondents score based on the answer key with minimal error. Basically for each question I wanted to have a drop down box with the three choices and then have the numeric code automatically calculated in a separate record in the table. I can't seem to figure it out. Please help!! |
#3
|
|||
|
|||
Likert Scale data- how to enter in Access
Thanks so much your response.
So what exactly does "committing spreadsheet" mean? Yes, you hit the nail on the head- I currently have a single record for each respondent, and each response is a column. How should I restructure the data? Have each response as a separate record? I certainly don't want to commit spreadsheet, whatever it is! As I mentioned in the original post, the Likert values are in a separate table, where each record is a question and stores the value assigned to each response category (ie if "I agree"= 3 or if "I agree"=1). Maybe restructuring my data will make the relationship that much clearer. Thanks "Dale Fye" wrote: Well, from the sounds of your post, it sounds like you have a single record for each respondant, and a column for each response. Is that correct? If so, then you are "committing spreadsheet" and need to restructure your data. Post back with the structure of your Response and LikertValues tables and I'm sure someone will be able to help you. For a good tutorial on surveys, take a look at: http://www.rogersaccesslibrary.com/O...p#Hookom,Duane. "Struggling graduate researcher" Struggling graduate wrote in message news I have created a database to record responses to a questionnaire in which respondents must answer YES, NO or I Don't know for each question. For EACH question, their response gets a different weight- for example, sometimes a YES response = 3, sometimes = 1. SO I have created an "answer code table" in which each question has a unique ID, and then I entered the correct values for YES, NO, and I DON"T KNOW responses. Then I created a "survey" table record responses, where each survey has a unique ID. SO now what I want to do, is link the answer key table with each survey response, so that I can easily enter the survey data and calculate each respondents score based on the answer key with minimal error. Basically for each question I wanted to have a drop down box with the three choices and then have the numeric code automatically calculated in a separate record in the table. I can't seem to figure it out. Please help!! |
#4
|
|||
|
|||
Likert Scale data- how to enter in Access
"Committing spreasheet" is creating column names that are actually data
points (in your case survey item #s). A lot of times, people who are used to working with spreadsheets just assume databases should be configured the same way, but to take advantage of the relational aspects of a database, you have to turn that on its head. Another good example is budgeting. Many people create spreadsheets that have expenditure categories down the left column of a spreadsheet, and have months listed across the top, and they enter either estimated or actual expenses in the cells of the matrix. In a relational database, you would generally set this up as 3 columns (ExpenseType, ExpenseDate, and Amount). Then you can use queries to present the data in a wide variety of formats (including a crosstab query, which looks like your spreadsheet). If done properly, your survey responses will go in a table that looks something like: RespondantID, QuestionNum, Response Then, your LikertScale table would have columns like" QuestionNum, Response, ResponseValue Then you can create a query: SELECT RespondantID, SUM(ResponseValue) as CumScore FROM tblResponses INNER JOIN tblLikertScale ON tblResponses.QuestionNum = tblLikertScale.QuestionNum AND tblResponses.Response = tblLikertScale.Response GROUP BY tblResponses.RespondantID This would give you each respondants ID and their cumulative score. Or you could create a query: SELECT QuestionNum, Response, Count(RespondantID) as Frequency FROM tblResponses GROUP BY QuestionNum, Response To get a quick count of the number of each response to each question. To do this with your "spreadsheet" structure would be much, much more difficult. HTH Dale "Struggling graduate researcher" t.com wrote in message ... Thanks so much your response. So what exactly does "committing spreadsheet" mean? Yes, you hit the nail on the head- I currently have a single record for each respondent, and each response is a column. How should I restructure the data? Have each response as a separate record? I certainly don't want to commit spreadsheet, whatever it is! As I mentioned in the original post, the Likert values are in a separate table, where each record is a question and stores the value assigned to each response category (ie if "I agree"= 3 or if "I agree"=1). Maybe restructuring my data will make the relationship that much clearer. Thanks "Dale Fye" wrote: Well, from the sounds of your post, it sounds like you have a single record for each respondant, and a column for each response. Is that correct? If so, then you are "committing spreadsheet" and need to restructure your data. Post back with the structure of your Response and LikertValues tables and I'm sure someone will be able to help you. For a good tutorial on surveys, take a look at: http://www.rogersaccesslibrary.com/O...p#Hookom,Duane. "Struggling graduate researcher" Struggling graduate wrote in message news I have created a database to record responses to a questionnaire in which respondents must answer YES, NO or I Don't know for each question. For EACH question, their response gets a different weight- for example, sometimes a YES response = 3, sometimes = 1. SO I have created an "answer code table" in which each question has a unique ID, and then I entered the correct values for YES, NO, and I DON"T KNOW responses. Then I created a "survey" table record responses, where each survey has a unique ID. SO now what I want to do, is link the answer key table with each survey response, so that I can easily enter the survey data and calculate each respondents score based on the answer key with minimal error. Basically for each question I wanted to have a drop down box with the three choices and then have the numeric code automatically calculated in a separate record in the table. I can't seem to figure it out. Please help!! |
#5
|
|||
|
|||
Likert Scale data- how to enter in Access
Thank again- currently restructuring the data as you recommended. HOwever, I
am still confused by your recommendation for the LikertTable structure. IF I just have questionNum, Response, and responseValue as my three fields- what I need to do is have a lookup box for response- with YES, NO, and Dont Know- and then is there a way to create an expression for EACH question- "if YES" then 3, "if No" =1? This is an important point, as I said, not all questions carry the same value for YES. Ex. for QuestionNum 1, if YES=1, if DONT KNOW=2, if NO=3, but for QuestionNum2, if YES=3, if Don't Know= 2, if NO=1. Also, I would like to have the ResponseValue field in the survey response table, so that the VALUE also appears. I am not as much interested in finding the cumulative score, but rather making it easy for someone to enter the data- so that they just choose in the look-up table YES, NO, Don't KNOW- without worrying about if YES=3 or if YES=1. I want it to automatically come up. THanks for your patience "Dale Fye" wrote: "Committing spreasheet" is creating column names that are actually data points (in your case survey item #s). A lot of times, people who are used to working with spreadsheets just assume databases should be configured the same way, but to take advantage of the relational aspects of a database, you have to turn that on its head. Another good example is budgeting. Many people create spreadsheets that have expenditure categories down the left column of a spreadsheet, and have months listed across the top, and they enter either estimated or actual expenses in the cells of the matrix. In a relational database, you would generally set this up as 3 columns (ExpenseType, ExpenseDate, and Amount). Then you can use queries to present the data in a wide variety of formats (including a crosstab query, which looks like your spreadsheet). If done properly, your survey responses will go in a table that looks something like: RespondantID, QuestionNum, Response Then, your LikertScale table would have columns like" QuestionNum, Response, ResponseValue Then you can create a query: SELECT RespondantID, SUM(ResponseValue) as CumScore FROM tblResponses INNER JOIN tblLikertScale ON tblResponses.QuestionNum = tblLikertScale.QuestionNum AND tblResponses.Response = tblLikertScale.Response GROUP BY tblResponses.RespondantID This would give you each respondants ID and their cumulative score. Or you could create a query: SELECT QuestionNum, Response, Count(RespondantID) as Frequency FROM tblResponses GROUP BY QuestionNum, Response To get a quick count of the number of each response to each question. To do this with your "spreadsheet" structure would be much, much more difficult. HTH Dale "Struggling graduate researcher" t.com wrote in message ... Thanks so much your response. So what exactly does "committing spreadsheet" mean? Yes, you hit the nail on the head- I currently have a single record for each respondent, and each response is a column. How should I restructure the data? Have each response as a separate record? I certainly don't want to commit spreadsheet, whatever it is! As I mentioned in the original post, the Likert values are in a separate table, where each record is a question and stores the value assigned to each response category (ie if "I agree"= 3 or if "I agree"=1). Maybe restructuring my data will make the relationship that much clearer. Thanks "Dale Fye" wrote: Well, from the sounds of your post, it sounds like you have a single record for each respondant, and a column for each response. Is that correct? If so, then you are "committing spreadsheet" and need to restructure your data. Post back with the structure of your Response and LikertValues tables and I'm sure someone will be able to help you. For a good tutorial on surveys, take a look at: http://www.rogersaccesslibrary.com/O...p#Hookom,Duane. "Struggling graduate researcher" Struggling graduate wrote in message news I have created a database to record responses to a questionnaire in which respondents must answer YES, NO or I Don't know for each question. For EACH question, their response gets a different weight- for example, sometimes a YES response = 3, sometimes = 1. SO I have created an "answer code table" in which each question has a unique ID, and then I entered the correct values for YES, NO, and I DON"T KNOW responses. Then I created a "survey" table record responses, where each survey has a unique ID. SO now what I want to do, is link the answer key table with each survey response, so that I can easily enter the survey data and calculate each respondents score based on the answer key with minimal error. Basically for each question I wanted to have a drop down box with the three choices and then have the numeric code automatically calculated in a separate record in the table. I can't seem to figure it out. Please help!! |
Thread Tools | |
Display Modes | |
|
|