I have the following scenario:

I'm in the filed with printed versions of my survey. I would like my research assistant to do the first version of the transcription. Therefore I would like to let him use the online-survey since all data will then be automatically inserted in the database and it seems to me more efficient and reliable than creating an excel file or similar.

After his transcription I would like to check the information he entered and if necessary change answers or complete missing information. Furthermore I would need to edit some of the information at a later point in time. The printed survey will be taken out for about 120 respondents.

Is there an efficient way to organize this? I know there is the "Allow editing answers after completion" option but than how do I manage the different answers? Would it be needed to create 120 Users with username or how do I achieve this?

Thx in advance!

Hi,

sorry, but using limesurvey for this scenario seems for me an unusual use.

Limesurvey is a really nice tool to collect data and export it. Which advantage offers limesurvey for transcribing?
Okay, once you set up the survey, you can put in data and export it as usual with the advantage of a nice syntax e.g. for SPSS. For easier input you can use the "data entry screen" ( www.limesurvey.org/manual/Data_entry ). But editing, manipulating and complete each data should be easier in excel.

In your case we don't use limesurvey, we code the answers of a printed survey and transcribe it in excel. Import the data in SPSS and write the syntax with all the labels and values. If necessary we edit the data in duplicated excel-file, so we keep the originals. Logical manipulating like indices or renaming/recoding we make in SPSS.

I might be wrong, but for me the effoert of setting up the survey and hacking in all data would exceed the old fashioned way.

Greets
Stefan

Hello Stefan,

thx for your answer. Just in case, is there not any software that is more efficient in doing printed interviews and transcribing them than the old fashioned combination of MS Word and Excel? Sine the survey is evolving (you might eventually ask an additional question that has not been asked before), I have to redo manually the transcription sheet all the time and afterwards it will be chaotic bringing all the information together...

A quantitative analyse should be standardized, so you usually don't have to add questions. Adding questions to limesurvey while the survey is online (while you input data) doesn't work anyway. Once you add data you can't add a question without dropping the other answers, because the database has to be reinitialized.

Maybe it is possible to wait until the 120 participants answerd, then you can see, wich questions you have to add, and then starting to input the data ("data entry screen"). The editing of the data would then work under "responses and statistics" --> "display responses".

I agree with Steve. If the survey is evolving over time you'll have big problems with data entry anyway.

We use Limesurvey for data entry of offline questionnaires. But these questionnaires are standardized questionnaires. You don't add or delete questions (otherwise it will be difficult to compare anyway).

However, transcribing sounds rather like qualitative fieldwork, than standarized quantitative research?

By transcribing I refer to converting the written information to a digital form and yes, of course ideally the questionnaire is perfect before you go to the field. Still, in reality questionnaires have to be adapted and if you don't want to loose the information from the first interviews at all you will have to include them (even with missing information on eventually included extra questions).

Still the main issue is about the question if I can edit/review a first data-entry again if necessary. But if it is all to complicated I will stick to the Word/Excel combination...

You can go into each questionnaire and edit responses.It is not complicated at all.

The question is, are all those 120 interviews typed at the end, when you have the final questionnaire version or not? Because when Limesurvey is activated, you only can make certain changes (from what I know you can't add answer options, etc.). To include new questions, etc. you would have to deactivate the questionnaire which means the answers given so far are gone (download them before you deactive). In some cases, you can reimport these previous answers into Limesurvey again, but I don't know if you can do this if there have been made certain changes to the questionnaire (like new questions, new answer items, etc.). You would need to test that out (or maybe someone else has a better answer).

However, I would always prefer Limesurvey over Excel for data entry of longer questionnaires. But the editing I would always do in Excel because it should be so much quicker.

My approach would be:

1. Create a questionnaire that is as good as it can be.
2. Start data entry. Try to avoid changes to the programmed questionnaire as much as possible. If you need to make changes, export the results typed so far to Excel. Then deactivate the survey, make the changes, activate it again. See if you can import the previous data. If yes - great. If not, you still have it in a pretty standarized format in Excel.
3. Go on with data entry.
4. At the end, export everything to excel. If necessary include the other files to final excel and adapt the columns accordingly (should be quite quick.

Then I would revise things in Excel. It is probably so much quicker. You can do it in Limesurvey (only in the case that you could get old answers back into the system). But I think it would be a lot slower.

Sounds like a good strategy.
The data entry should take place right at the day after each interview and the revision one day after the first data entry. The only thing I could not figure out is, how can I edit the first data entry on the following day?

You go to your survey and under survey responses you can view the response (magnifying glass) or edit (pencil).

got it. thanks for your help!!