malambomutila/Airtel-Employee-Retention-Survey

Data analysis on employee turnover and retention in the telecommunications services sector in Zambia. The data used was from an online survey conducted in Zambia by Mwila (2023). # Employee Retention Survey at Airtel Zambia 2023 This analysis reproduces the results obtained by Mwila (2023) in a study on employee turnover and retention in the telecommunications services sector in Zambia. The data used was from an online survey conducted by Mwila (2023) involving employees at Airtel Zambia. This analysis focuses on age and employee retention. ### Aim To use Python to reproduce survey analysis results obtained in SPSS. ### Guiding Questions To reproduce the results obtained by the Mwila (2023). The goal of this data analysis was to answer the following questions: - What factors contribute to high employee turnover at Airtel Networks Zambia? - Do policies impact the reduction of staff turnover and improve performance at Airtel Networks? - What are the differences in turnover rates among various age groups of employees at Airtel Networks? ### Methodology 1. Data Cleaning 2. Reliability Test 3. Descriptive Statistics ### Data Cleaning Firstly, the raw data was read into a Pandas dataframe, containing columns named Timestamp and Score, which were subsequently dropped from the dataset. Following this, uniform naming conventions were applied to the columns. Next, the data was split into quantitative and qualitative subsets. The last four qualitative questions were dropped to isolate the quantitative data. Subsequently, the columns were renamed to ensure consistency and clarity. Value labels where assigned and a glossary of variables was then created to document the variable names alongside their corresponding original questions or descriptions. Further cleaning involved dropping rows with missing values and replacing specific responses to standardise the data. Glossary obtained by Mwila (2023) in SPSS Glossary obtained using Python ### Reliability Test Cronbach's Alpha was used to test the internal consistency and reliability of the questionnaire and the quantitative results. The Cronbach Alpha values are in the range 0 – 1.0. The value …