Based on feedback from the professor and from your reading/experience, present an upgraded version of the research problem, research questions, and hypotheses.

In the previous work, the proposed work was related to the healthcare data and the professional knowledge. The information is processed through the validated process and can be considered for the future perspective of diagnoses of disease. The diagnosis process depends on the use of machines that can be tracked by professional doctors. The previous work provides a complete research plan for the operational model. The research questions are perfectly aligned with the considered variables, relation, and statistical analysis of the research.

The research questions considered in the analysis were related to the contribution of data science, widespread diseases for the diagnosing process, influenza treatment, and use of science in the identification of risk factors and other treatment processes. The research can be updated for the operational model analysis. The research problem is specifically related to the proposed plan and depends on the observation’s goals and objectives of the research. The research questions describe problems, variables, and their relation.

The statistical analysis is considered in the analysis. There is two proposed hypothesis including the null hypothesis and alternative hypothesis. The hypothesis deals with the significance of relation of data science with the tracking, curing, and diagnosing of widespread diseases. The alternative hypothesis considers data science for tracking, diagnosing, and widespread diseases. The impact of data science is also considered in the hypothesis.

The updated version of research is based on the operational model and investigation of data science role in the treatment of diseases. In the latest updated hypothesis, the more precise statement is considered that defines the relation between the treatment process and data science. The role of new technology and machines is significant in disease analysis, diagnostic understanding, and tracking of issues of disease.

1.      Discuss possible sources of uncertainty: sampling error, researcher bias, reliability and validity of the instrument.

There are different sources of errors and uncertainty in the research and these sources of error include sampling error, researcher bias, the validity of the instrument, and reliability. The sources of error can be due to faulty recording and the wrong recording of the measurement (Statistics. laerd. com, 2018). The misreading of scale can cause blunders in research. The errors can be classified into three types such as systematic error, random error, and blunders. The systematic errors can be identified on the basis of sources and causes (Ward, Self, & Froehle, 2015).

 The data errors can be due to the wrong substitution, systematic biases, random biases, missing and substitution (Statistics. laerd. com, 2018). The small data errors induce measurable effects on the goals of operational performance. The systematic substitution errors result as an increase in the frequency of errors and the shorter duration metrices of the results can decrease performance issues. The metrices of longer time duration have a lower proportion of data errors (Ward, Self, & Froehle, 2015). The optimization of instruments is required to enhance the reliability and validity of the instrument. The impact of errors on the operational performance can be identified and then measured at a low and higher frequency of data. The perception of the operational model requires to underline the metrics and potential impact of operational data on the analysis (Statistics. laerd. com, 2018).

2.      Assume one of the hypotheses is true, and your study produces that result. What does that mean to your study?

The level of significance demonstrates the probability of accepting and rejecting the null hypothesis if it is true (Statistics. laerd. com, 2018). In the present study, the null hypothesis is "data science has no significant relationship with day to day living" and the alternative hypothesis is "data science has a positive impact on day to day living. Consider if the significance level of the hypothesis is 0.05 that indicates 5% of the risk in concluding the existing and actual differences. If the value of “p” is less than or sometimes equal to the level of significance than we can reject the null hypothesis (Statistics. laerd. com, 2018).

The statistical analysis provides vision about the rejection of the region, If the value of p is smaller than the significance value than it can be rejected. In the present consideration if we reject the null hypothesis that it measures the alternative hypothesis is acceptable (Ward, Self, & Froehle, 2015). On the basis of selected and rejected hypothesis, it can be concluded that data science has a positive impact on the day to day living. The probability of rejecting the hypothesis is different in both cases including null and alternative hypothesis (Ward, Self, & Froehle, 2015).

3.      Assume one of the hypotheses is true, but your study causes you to reject that hypothesis. What does that mean to your study?

In the previous assessment two hypothesis were proposed including null and alternative hypothesis the n Null Hypothesis states that “Data science has no significant relationship with diagnosing, tracking and curing some of the world’s deadliest and widespread diseases” while on the other hand the Alternative Hypothesis states that Data science has a positive impact on diagnosing, tracking and curing some of the world’s deadliest and widespread diseases (Ward, Self, & Froehle, 2015).

There are some alternative solutions for the condition of true hypothesis and if the study rejects that hypothesis. Consider if the level of significance is above then the cut off value than the null hypothesis cannot be rejected and at the same time the alternative hypothesis cannot be accepted (Ward, Self, & Froehle, 2015).

The hypothesis can be considered correct if the study also matches with the results (Ward, Self, & Froehle, 2015). On the other hand, if the statistical analysis shows that the level of significance below the cutoff value then alternative hypothesis should be accepted. Basically, the cutoff point describes the statistical analysis and stimulates the hypothesis as accepted or rejected (Ward, Self, & Froehle, 2015).

4.      Suppose the results are statistically significant (p < 0.05 and the null hypothesis is rejected), but the effect size is very small. How would that influence your interpretation?

According to Cohen the smaller effect size is approximately equal to 0.2 and the medium effect size is 0.5 and the larger effect size is 0.8. The mean of the two groups can be different and it can be different by 0.2 or more. The difference can be trivial and statistically significant. The interpretation of smaller size is different and attributes to the unexplained variance for the dependent and independent variance (Ward, Self, & Froehle, 2015).