
Transcript 2
The main difference between primary data that you collect yourself and secondary is generally you are much more restricted in what you can collect if you are collecting your own data is simply because of the level of resources that's available to you so
I created a dataset along with some colleagues on content analysis of data on newspapers. That was relatively easy to do because this is just a question of buying newspapers and carrying out the analysis, doing the counting and so on. It didn't involve for example travelling to interview people things like that.
In secondary data analysis it's possibly used techniques like computer-assisted personal interviewing whereby a hand-held computer guides interviewer through the routine of questions that can be used for different kinds of respondent on a survey.
The kind of resources that can go in into the creation of secondary data are usually much larger than the level of resources that an individual researcher will have at his or her disposal in a university and some more and more quantitative research uses secondary data just because the quality of the data produced his much higher. It's based on very well drawn random samples that can take advantage of things like stratified sampling or cluster sampling to maximize the value of the data that's gathered.
Cognitive testing is done of the questions used to try and make sure that questions are actually measuring what you hope they are measuring, the data is properly cleaned and now because of the technological advances in information technology of the last ten fifteen years means that centrally created data is much more easy to disseminate and get access to from individual universities that's no longer a question for example, of sending away for computer tapes and things like that. You can go on the web and download a dataset in a matter of minutes.
Writing a data management plan by Edina, University of Edinburgh, modified by Glasgow Caledonian University is licensed under a Creative Commons Attribution 3.0 Unported License. Based on a work at http://datalib.edina.ac.uk/mantra/introduction.html.