Research Help: NURS 3710 (RN to BS): Evidence-based Practice in Nursing: Primary Research vs. Secondary Research

Primary Research vs. Secondary Research: Why does it matter? How do I tell which one I'm looking at?

Different databases may contain different types of sources. When you are conducting research you need to consider what types are most likely to contain the type of information you need to answer your research question. Depending on the database you might find books, magazine articles, trade publication articles, scholarly journal articles, documents, newspapers, videos, audio clips, images, et cetera. For the purposes of your Nursing research we will focus on sources that are primary research studies. Before you can do that, you need to know the difference between primary research and secondary research, how to identify which one you are looking at when you evaluate a source that you find and why the emphasis gets placed on using primary research studies.

Primary Research: This is research that is done by the author of the source you are using where that author conducted some method of research to gather new data that s/he then reports, analyzes and interprets in that source. Primary = original, first-hand; the author of the source generated the research data they are using.

Secondary Research: This is when an author of the source you are using gathers existing data, usually produced by someone else, and they then report, analyze or interpret that other person's data. Secondary = second-hand; the author of the source did not generate the research data s/he is using.

Note: In some fields, such as nursing and medicine, the line between primary research and secondary research can blur. Well done systematic reviews and meta-anaylsis, while using data generated from other peoples' research, can sometimes be used as primary research to identify patterns and trends. The key to a well done one is being able to find a large enough number of research studies that use the same methodology/protocols to measure/assess the same thing so that the researcher doing the systematic review and/or meta-anaylsis is comparing x to x.

One example of well done systematic review and/or meta-anaylsis would be a researcher who wants to compare the median pay and entry-level education for nurses over the past thirty years. Instead of conducting his own survey he uses the data gathered by the Bureau of Labor Statistics published in each of their annual Occupational Outlook Handbook published during that time. He then applies meta-analysis to test the pooled data to see if salaries increased when a higher education level was required. Why is this good? The original data being used is from a national study conducted each year using the same methodology by the same institution with the same data points so that across that time period so he can be relatively secure that he is comparing the same thing (median salary and education level) year-to-year.
One example of a poorly done systematic review and/or meta-anaylsis would be a researcher who has the same research question as above, but selects thirty different studies that were each conducted only once at some point during that time period that were done by different researchers using different methodologies. Some of the studies reported the mean salary while others reported the median. Some of the researchers gathered the data by querying nurses working in the field, others queried only hospital personnel departments, while others queried college career placement departments. Some studies where national in scope while others where international, regional or state-based. The researcher than applies meta-analysis to the salary data and reports that nursing salaries increased by 2% each year during the 1990's, 1% per year between 2000-2007 and by 2.5% per year since 2008. How reliable is his analysis? The mean and the median of a dataset can be very close to each other or they can be really far apart so in essence using studies that reported the salary differently would be like comparing an apple to an orange. The same would be true of major differences in how the data was gathered in this case.

Elements of a Primary Research Source	Elements of a Secondary Research Source
Primary (original) research can be in the form of an: experiment survey case study focus group interviews The key is that it generates some type of data that the researcher can then analyze and utilize to prove/disprove their research question.	Secondary research is usually in the form of a review. There are different types of reviews: Literature Review - Provides a representative collection of the primary research studies that have previously been conducted by others usually within a specified time period such as the last 10 years to try to answer his/her own research question without doing primary research himself/herself. The author summarize commonalities, differences and identify holes where more research is needed through his/her interpretation of other researchers' data. Systematic Review - Is an expanded literature review that tries to collect and summarize ALL of the primary research studies that have previously been conducted by others as they try to answer their own research question. Meta-analysis - Is when the author of a systematic review uses statistical methods to summarize the results of the data from the studies s/he found. No new data is being generated because of primary research - the author of the meta-analysis is using statistics to enhance her/his analysis of other researchers' data. There are always exceptions, and one of those is that a well-done meta-analysis can generate new data in the form of identifying patterns and trends. It is also often used in other types of resources such as non-research based articles, books, documentaries, et cetera to provide background information that ties back to primary research that the person using the source can locate if they want to see the actual primary research study.
Author is the researcher and conducts an original study gathering the data that results from that study.	Author gathers only research studies and data that were generated by other researchers.
Author will include a section that includes details about the research methods used, how data was gathered, participants, et cetera.	Author includes little or no details about the research methodology used by the original researchers of the studies they used.
Author provides the reader with the data in the results section.	Author may generalize the collective results and/or only highlight the results they felt were important. You don't have access to the original data to determine if the generalizations are accurate and/or if results key to your research question were left out.
Author begins the research study with a brief overview of previous research on the topic and relates where the study they are conducting fits into that scholarly conversation.	With a secondary research source you only have the current author's context for how the selected research studies connect into the scholarly conversation on the topic. His/her context may differ from the context of one or more of the original researchers, but you won't know this unless you track down the original studies.
The reader of a primary research study can use the information provided in the methodology, analysis and results sections to help judge the quality of the study.	The reader of a secondary resource has no way to judge the quality of the research studies selected by the author unless the reader tracks down the original study.
The reader, having access to the data, can analyze and interpret that data in context with their own research question.	The further away the reader gets from the original research source, the more likely the reader will lose or misinterpret the original context of the data being used. For example: I publish a research study (primary research) in 2014 that is focused on parent monitoring of there child's cell phone usage. In my study I said I surveyed 800 ninth graders in Massachusetts and that data I gathered for a specific question I asked showed that 10% don't have a cell phone, 70% have a cell phone on their parents plan and 20% have a cell phone on their own individual plan. Another researcher uses my research in their literature review published in 2018 and summarizes my findings as "one survey found that 90% of ninth graders have cell phones". You find the 2018 article and generalize that author's summary of my findings to state that "a 2014 survey found that 10% of ninth graders don't have a cell phone". How accurate is the context in which you interpret my data? How accurate is the context of the data for the person who reads your literature review? By the time you are using my data from the secondary resource neither you or your reader have any idea that this was a study of 800 participants, that it was focused in only one state, or that it included data for those with a phone as to whether they have their own phone plan or on their parents' plan.

Tips when first evaluating a source: Once you find a research-based source, read the abstract and/or methodologies section and ask yourself who conducted the actual research process to gather the data?

If the author(s) indicate they gathered the data first-hand by surveying a specific population, creating and running an experiment, conducting a focus group, observing a specific population/task, et cetera, they are doing primary research.
If the author(s) indicate that they used only data gathered from other people's research studies by reviewing the literature or research and aren't generating any new data, they are doing secondary research.

Common sources where primary research in the field of Nursing is published are:

Scholarly/academic journals - these are journals that are published by academic publishers (colleges, universities, et cetera) and professional organizations.
Conference Proceedings
Books
Dissertations
District, State or National Reports

Any of the above sources may also be peer-reviewed, meaning that the content is reviewed by other professionals in the field before it is published. Peer-reviewed, scholarly journals are considered to be high quality and often are a requirement in your research assignments because they are produced by experts and professionals in the field and all primary research articles are put through a peer-reviewed vetting process that is detailed by the journal publisher. It is also where professionals in the field in turn tend to publish their research for those same reasons. It is important to keep a couple of things in mind:

Not all scholarly/academic journals are peer-reviewed.
Not every article in a peer-reviewed journal is a primary research article or even a secondary research article. These journals often contain book/product reviews, opinion pieces, advice for practitioners, literature reviews, et cetera.
Books take longer to publish and when you need current research/data this is a consideration. It can also be hard to determine what, if any type of peer-review process a book has gone through.
Dissertations are by their very nature peer-reviewed, especially at the doctoral level, but can be hard to access as they are often only available at the university where the dissertation was written.

Research Databases are a great tool for finding and accessing primary research articles published in scholarly/academic, peer-reviewed journals. In the field of Nursing our library provides access to CINAHL Complete, Cochrane Collection and Proquest Nursing & Allied Health. Each of these databases have search functions to help you narrow your results list down to scholarly/academic, peer-reviewed journals.

Depending on your topic and research question you may need to explore research databases in other disciplines as well. Here are two examples:

How does lead poisoning impact the cognitive development of children? In edition to nursing/medical databases I might want to also look at psychology and biology resources.
What impact does arts therapy have on children with behavioral issues in the classroom? In edition to nursing databases I might want to also look at psychology and education databases.