REVIEW: DATA ANALYSIS SOFTWARES AVAILABLE FOR RESEARCHERS
By Samuel EkpoTweet
During or after research, there is usually a need to analyze; to present the information gotten in a presentable and easily understood format.
There are numerous packages or softwares that could be used to achieve this.
The choice of the method to use in the analysis depends on the function to be achieved. However, even for the same function, there are multiple means of analysing data, hence, certain things should be noted.
The method might be based on the popularity of the option in your environ. It might be based also on what you were taught with in school. It might be specially required for the kind of work you are doing. Possible options are discussed below. All options have their pros and cons.
- Excel: This is a common medium. It is part of the Microsoft Office packages. Aside MS word, it is easily the most used MS office package, before PowerPoint.
The major advantage is that it is easy to use. It is cheap to install and readily available. You rarely get stuck in its use due to its widespread popularity. It is used for general analysis: regression, scatter plots, graphs and various charts, various functions, visual presentation of tables et al. Wide applications in the business area as well. It is not programming -oriented, rather it is analytical. However, it is belittled in advanced research and difficult as well as slow for large data . Seen as elementary.
- SPSS and strata: This is common in science. They are specific for analysis and can do better in that regard. There is a possibility of merging datasets together as well. The various sheets in excel can be exported and merged elsewhere.
Statisticians and some social scientists use it as well. Stata is easier and cheaper to use than SPSS. SPSS would most likely be preferred of the two. They are both analytical tools, they don't require programming skills. Excellent for ANOVA analysis, t and chi square analysis, multiple regression analysis etc.
- Matlab: Full meaning: Matrix laboratory. This is common in engineering and scientific communities. It has good matrix support but lesser statistical analysis support. It is expensive and requires programming capabilities. Great for developing mathematical algorithms.
These are the few i have experience in. There are several others available. MatCAD is another powerful tool with lots of engineering applications.
SAS is another stat tool like SPSS though with very poor reviews due to its limits and use. Its distinctive quality is its large dataset. The range is in GB and might not be fitted at once. The learning curve is pretty high as well.
Then there's R, Scipy, PolyMath and the others.
Whichever option is chosen for analysis, consideration of the intended function, limits and alternatives is critical as it would save the researcher a whole of stress in the course of the analysis.