Its important to understand the different factors that work together to contribute to a successful project. This post will provide an overview of primary and secondary research methods, and I’ll explore this a bit more in the following few posts. Today, in particular, I’ll explain what primary and secondary research methods are, look at qualitative and quantitative research and I’ll also talk about triangulation. Before finishing the post, I will look at research strategies, thinking about your population and sample sizes. These are all things you should include when considering your own research approach.
Primary or Secondary Research
When working on big research projects we need to do both primary and secondary research. Despite the name, we do the secondary research first.
This is the material we read which has already been published: books, articles, items on the internet etc. If you access a resource to read something (this post for example), it would be classed as secondary research.
If you’re working on a research project, you will want to read secondary resources to get an understanding of your key themes. My literature review post go into this in more detail, but you want to ensure you are reading quality materials. Students sometimes show a tendency to rely on textbooks, but I would urge you to use these as background reading only (in a bibliography). In your work you should be using a range of academic journal articles, government and other official publications, surveys, research papers, and any relevant trade and professional body publications. You should also only use resources that you trust and are credible. You might find this series helpful.
Once you’ve completed your secondary research and written your literature review you should have identified your main themes and the research questions you want to answer. You’re now ready to move onto your primary research.
You have done your reading and you have decided to issue questionnaires to find answers, this is your primary research. It is work which has not been done before, has not been analysed or published yet. You send out your questionnaires, gather them back in and analyse the results, and finally write up a report presenting your findings. This is your primary research, but if someone else then picks up and uses your report in their work, then its their secondary research.
For research reports, its important to do both primary and secondary research. Primary research allows you to seek the answers to your questions or test the hypotheses arising from your secondary research, which is why we do it first.
There are a range of different types of primary research that you can carry out, including questionnaires, interviews, observations but I will look at these in a future post. Now, we will explore qualitative and quantitative research.
Qualitative and Quantitative Research
Ideally we want a mix of both qualitative and quantitative data methods so you can provide facts and figures – numbers (quantitative) and the explanations behind the statistics (qualitative). Different primary research methods will provide either qualitative or quantitative results, and some will provide both. Questionnaires with closed questions or rating scales will provide quantitative results – how many people said yes, or on a scale of 1-5, what proportion of people selected 1, 2 etc. These are quantitative answers. If questionnaires have an open question (please explain why you gave the answer you did) then that is allowing for a more detailed, richer answer which we would classify as qualitative as it adds quality to the statistics we have. Interviews, for example, would usually provide qualitative answers as we are able to ask for more information, making an interview much richer.
Secondary research should also contain both quantitative and qualitative information.
You can read more about the difference between qualitative and quantitative research in this post.
This is the final stage of your research. Triangulation means that you’re gathering the information from more than one source. Usually this would be to check that the results are duplicated, but it also allows you to find more information. For example, you might send out questionnaires to all staff in your organisation, but conduct follow up interviews to ask questions that have arisen from the analysis of the questionnaires. Alternatively, the same questions could be sent to two (or more) different groups of people to confirm understanding. For example, asking employees and their managers about the role of the manager – do the employees and the managers have the same ideas, or is there some conflict there. Triangulation allows you to identify where disagreements, discrepancies or conflicts may be present. Triangulation allows you to test your research and to improve its validity.
Once you’ve decided on your research methods, ensuring a range of qualitative and quantitative methods and applying some triangulation, you also need to think about how many people should be involved in your research.
This is the number of people in the group you are studying. If its your workplace, this could be all the staff, or if you’re only looking at one area, one site or department then the population would be everyone in that department. It could equally be a group of employees – professional staff, administrative staff etc. You need to identify who is going to be represented by your research and that is your population – this is not the same as the people who are going to participate.
Whatever you decide is your population, it needs to make sense in the context of your research. If you’re working on a piece of academic research, your supervisor should be able to guide you more here.
Once the population has been identified, you need to think about how many people are going to be participating (the participants or respondents). If its a questionnaire, it could be 100% of your population, or you could decide to send the questionnaire to a representative sample of 50%. There are different ways of choosing your sample, which I’ll cover in another post, but you should bear in mind that the bigger your sample, the more accurate and reliable your results will be.
Your sample should be broad enough to be representative of the full population. If your population is 10,000 then 10% would still give you 1,000 questionnaires returned. But if your population is only 50, then 10% would only give you 5 questionnaires, so you would need to think about whether that is going to be enough to speak on behalf of the whole population. For example, if your government did a survey representing the whole population of your country – would you feel that only getting answers from 100 people would be enough to speak for the whole population? Or if they only sent questionnaires to people living in the capital city? You need to ensure your sample is representative of your population.
This is another figure you need to consider. It will be highly unlikely you will get 100% of the questionnaires back from your sample. If you send out 1,000 questionnaires, the response rate is the percentage of those that are returned. The response rate is calculated from the sample size, not the population. Again, you will want to think about how reliable the response rate is. If you only have a 10% response rate, you would need to think about how representative this can be of the population. You should carry out your analysis based on the results you have back, but you should acknowledge the impact a low response rate (or a high rate if you’ve been lucky) on your work. Again, if you have a supervisor, they should advise you more specifically here.
When writing up your methodology section you will be outlining what you’re trying to find out, what primary and secondary methods you’re going to use and why. What obstacles you might encounter and how you will overcome them. You will also need to talk about your sampling strategy, including population, sample size and response rates.
I hope you have found this latest post in the Research Project Series helpful, but as ever, please let me know if you have any queries.
There will be further posts for this series, and more generally, I’ll be posting on study skills throughout the year, so if you’re a student looking to get the most out of your studies, sign up to my blog so you don’t miss out.