Step-By-Step Guide on Data Analyses for Academic Researchers

According to Wikipedia, data analysis involves the process of inspecting, cleaning, transforming, and modeling data with the goal of discovering useful information, suggesting conclusions and supporting decision-making. Data analysis has multiple facets and approaches, encompassing diverse techniques under a variety of names, in different business, science, and social science domains.

Further, The Business Dictionary defines data analyses as the process of evaluating data using analytical and logical reasoning to examine each component of the data provided. This form of analysis is just one of the many steps that must be completed when conducting a research experiment. Data from various sources is gathered, reviewed, and then analyzed to form some sort of finding or conclusion. 

Data analyses involve the following steps:
1) Statement of hypotheses
2) Specification of model
3) Obtaining/collecting data
4) Estimation/Analyses of Data
5) Testing the hypotheses
6) Forecasting or Prediction
7) Statement of policy implication/recommendations of results

The above details the steps involved in data analyses be it for an undergraduate project, post-graduate thesis or dissertation or a academic journal article for publication. Below, we provide details of what is involved and how each step can be achieved

Step one involves stating the actual hypotheses to be tested. Here, we assume that you have done your research to the stage were stating the actual hypotheses becomes necessary. I say this because your hypotheses flow from your identified research problem and objectives and research question. Also, you can only state valid hypotheses for your research because you have constructed your conceptual framework and studied the dependent and independent variables and related measurements and dimensions of your variable. 

The grasp of the issues surrounding the statement of hypotheses as enumerated above will prove to be very useful in the second step which involves Specification of model. The researcher can only specify the correct model because he/she has done adequate research as regards the theoretical and empirical relationships between the variables of the study. So understanding what previous authors and researchers have written in the topic area will prove very useful here. Your findings may end up contradicting the findings and postulations of previous authors and researchers but, you will probably need to base the method you use in your work on those previous researches.

In the third step, you collect the actual data you will be analyzing. This step may be as simple as clicking a few buttons if you are working with secondary data that is available on the internet. It may also be difficult and tasking if you are working with primary data.  So, if you have a choice of method, it is much easier and faster to work with secondary data. This is a good reason why it is very important to do a thorough back ground research before you even settle on any given topic. Collecting the data is actually only just a part of the activities in this step. The data collected also has to be: (a) Collated - into rows and columns (b) screened for defective, missing or incomplete data points. 

Step four involves estimating/analyzing the model as specified in step 2 above using the data collected, collated and screened in the step three. The method you choose in this step will be dependent on appropriate methods as proposed in previous empirical research or as agreed between the student and his/her supervisor(s). This step may also involve the use of some statistical software like E-views, SPSS, Stata etc depending on appropriateness and supervisor input. If you collected and collated your data correctly, this step should not be a big deal. You can easily visit the websites of any of the vendors of the software you intend to use for instructions. Some of the software also come preloaded with user instructions.

Step five involves the test of hypotheses stated in step one above but in most cases, this step is embedded in step four (Estimation/Analyses of Data). What remains to be done here is a sort of separation of the analyses results into their various parts. For example, some parts of the results will relate to given insight into the nature and structure of the data like: mean, Standard deviation, Variance etc. some parts will also be about diagnostics regarding the appropriateness of the data and the model(s) specified to test it. Some of the result outputs will also likely not be relevant to your study and specified model. Depending on the data and methods used, the tested hypotheses will not be too difficult to sieve from the mass of data in your results output.

In step six, the researcher has to provide an interpretation of the results garnered from data analyses step above. in interpreting the results, the researcher has to study closely data points that relates to the intended analyses. It is also important that the researcher study previous research results that used the same or related methods preferably from authors he/she already knows and can vouch for their expertise. It make also work in your favour to ask your supervisor recommend a few previous research of his or of other for to use as more or control to interpret you results. It is on the basis of the interpreted results that the researcher can actually forecast of predict the future outcomes.

Finally in step seven, the research on the basis of the results and interpretation in step six above proffer policy related solutions to the identified problems. For example, if the system is performing below expectation, how can it be remedied? If the system performing just fine, how can this be sustained and conceivably increased?. This will call into use all the background knowledge the researcher has gained in the process of carrying out the research. 

For more info on how to go about any part of your research work, drop a comment below or email us at optistrat.info@gmail.com.