NO52 Research Topic: Business Value and Challenges of Testing throughout Software development Life Cycle: A case from Sudan Software Market Research Hypothesis H1: There is a significant relationship between a poorly tested application and the number of bugs produced after development. H2: There is a significant relationship between testing and application performance. H3: There is a relationship between poorly tested application and the cost of maintenance of the application. Chosen approach to data analysis for the research question The aim of the study is to explore the impact of thorough testing on application performance, hence business performance. The approach i will use should enable me test the research hypothesis and present the result so that meaningful information can be drawn from it. According to Kerlinger and Lee, a "hypothesis is a conjectural statement, a tentative proposition about the relationship between two or more phenomena or variables and how they are to be tested" (2000 cited in Egbonwon, 2015:19). Therefore Univariate test approach will be used for my data analysis as it examines one variable per time across many cases (Easterby-Smith, 2012). It will be used to examine the effect of poor testing (independent variable) on application performance (dependent variable). I will start by using various statistics for describing, summarizing and presenting survey data. This approach enables me present the basic features of the data as simple summaries from where general information of the population can be drawn from (Easterby-Smith, 2012). Both mean and standard deviation will be used for data summary. How this data analysis approach will help answer the research question. Since Univariate test is used to determine the impact of a variable on another, and the kind of association a variable has on another. It help to determine if poor testing has a positive, negative or zero association (Easterby-Smith, 2012) with application performance. It will be used to test the association between poor testing and application performance, and the impact on business performance. This will be used not only to test the hypotheses of the study but also the research question which is 'What impact does a poorly tested application has on business performance'. How this data analysis approach is appropriate for the type of data i will be collecting and how it will ensure the validity and reliability (if appropriate) of your results. "Hypothesis testing allows researcher define how safe it is to go beyond a specific sample of data" (Easterby-Smith, 2012:261). This approach will help in effective testing of the research hypothesis so that general conclusion can be drawn from the result. The type of data i tend to collect from the survey includes the rate or frequency at which bugs are raised, amount of time spent on bug fixing, amount of resources spent on bug fixing, amount of effort, time spent on application support, and factors affecting the implementation of thorough testing. These data are significantly influenced by the quality of the application which depends on the quality of testing during and after application development. Therefore this approach will help in explaining the kind of relationship between these dependent variables and the independent variable (testing), hence how testing can influence their existence or impact on application. The data analysis and presentation approach used will impact on the research result positively. The analysis based on mean value aids the exhaustible use of sample data, it reduces bias in measurement and estimation and is more efficient than other summary measures of location (Easterby-Smith, 2012). Hypothesis testing gives safe boundary on how results can be generalized, increasing data reliability. Moreover, the use of bars and statistical values help to communicate results of the survey such that conclusion can be drawn from the sample data about the impact of the variables on business performance. Strategies to use to present the results of this type of data analysis. I will be using tables and graphs to show the frequency distribution of variables and various data. Mean values will be used for location measurement because of its efficiency as compared to mode and median values. Information about the central tendency indicates the variables (factors) having the highest impact on testing implementation. To calculate data spread, i will also use standard deviation which accounts for distance of scores from the mean. Also, the extent to which variation in one variable is related with variation in the other and the shape of association which could be positive, negative, curvilinear or linear (Easterby-Smith, 2012) could be analyzed with the correlation coefficient (r) and regression analysis. This will help to show how one internal factor influences the development of another. Challenges i foresee in interpreting and making inferences with this approach to data analysis. Analysis of data using mean value could present a challenge since it is not a robust tool for handling complex data due to bad inputs and large scores. Also the complexity of tools such as correlation coefficient could present a great challenge towards effective analysis. References Easterby-Smith, M., R. Thorpe, & P. Jackson, (2012) Management Research, 4th edition, London: SAGE Publications