Data analysis, an essential part of a dissertation, involves collecting & presenting data, interpreting the output and determining the pattern of the study. Data analysis process requires a huge amount of knowledge about various statistical tests, tools and approaches, without which performing data analysis can be a back-breaking task. Although cumbersome, data analysis is a rewarding process.
We work with SPSS mostly with research involving primary data collected through survey questionnaires.
Amos is used for structural equation modelling where a model’s fitness is to be checked/tested
Our team uses stata for research involving secondary data and where multi-variate regression and correlations, CFA and PCA are to be run
R/Python are used mostly for predictive analysis involving large data files.
The initial step involves collecting the relevant data. Depending on the study, data can be qualitative and quantitative. Various approaches such as surveys, focus groups, questionnaires, interview method etc. are used to collect the required data.
Not all the data that is collected is relevant to the study. This stage involves a thorough inspection of collected data and irrelevant data, white spaces, duplicate records, etc. is eliminated.
After data cleansing, the next step is the data analysis process. This step involves choosing a suitable statistical tool (SPSS, STATA, R language, etc.) and test (ANOVA, T-test, Z-test, etc.) to analyze the data.
The ultimate step is to interpret the obtained output and present the data in a readable format using charts, tables or graphs (pie chart, scatter plot, stem plot, etc.). By interpreting & observing the results, useful information can be discovered.