The Roles of Statistical Software in Data Analysis

Abstract

As quantitative research grows, application of statistical software (SS) becomes a more crucial part of data analysis. Researchers are experiencing a transition from manual analysis with paper to more efficient digital/electronic analysis with statistical software (SS). It identifies the prerequisites of producing world-class studies by using modern SS solutions. SS has contributed immensely in improving not only in demography studies and social investigation but also among other professionals’ researches in Nigeria and the world at large.A cross-sectional survey of 5 lecturers each were selected from 8 departments in the two faculties in Oye-Ekiti campus and were given questionnaire base on their availability and interest. A total sample size of forty (40) academic staff were selected but thirty (30) eventually responded which comprises fifteen less experienced staff with (0-5yrs) and 15 more experienced staff with (6yrs and above) as well as 10 non-response. The small sample size was due to few lecturers in some departments as at the time of this study. Data were analyzed using SPSS package.A univariate and bivariate analysis was done and findings of the study revealed that impact of statistical software on research results give Mean (M) =4.80 and Standard Deviation (SD)=0.41, on a (1-5) Likert scales with 80% Strongly Agree that SS has positive impact on their research result. Respondents category and running analysis without SS shows mean (M) =2.27 and Standard deviation (SD)=1.37 on a (1-5) Likert scales with 66% admitting that they cannot run analysis without SS. Some SS are suitable for some kind of analysis than others for instance, while SPSS,STATA, SAS MATLAB and R are 100% suitable for ANOVA, Eview, SAS STATA, R and MATLAb are 100% suitable for time series analysis. Furthermore, STATA, SAS, Eview, MATLAB and R are 100% suitable for different kind of regression analysis among others. In FUOYE, while SPSS has 92.9% knowledge and usage, others have usage and knowledge as follows; STATA (57.1%), SAS (15.4%), MiniTab (0%), Ms-Excel (76.9%), MATLAB (28.6%), R (0%), Epi-info (16.7%), and PSPP (8.3%). The paper concludes among other recommendations that academic staff/researchers should improve in their SS workshops training and further recommends integration of SS application in academic curriculum just like other compulsory courses. More so choosing statistical software packages to learn should be based on suitability of the software for all possible analysis you may wish to be analyzing.