1. Review and discuss type I and Type II errors associated with hypothesis testing.
2. Review and discuss the difference between statistical significance and practical significance.
3. Describe the common elements present in all hypothesis tests.

Hypothesis Testing

While testing the Hypothesis, a type 1 error occurs when a false positive result is obtained. Getting a false-positive result in the rejection of the null Hypothesis. The type 1 error occurs when data investigators reject the null Hypothesis (Crowder, 2017). However, in this incidence, the null Hypothesis rejected happens to be true to the population under study. The false-negative (Type II error) occurs when data analysts reject the HOwhose significance is essentially false to the population under study.

  Errors are common while conducting data analysis; this makes it difficult for investigators to avoid type 1 and Type 2 errors completely.  However, to increase accuracy and objectivity, using more extensive samples is the best thing to do (Emmert-Streib, & Dehmer, 2019).  The bigger the sampler space, the lesser the likelihood that results obtained will differ from the general population. False positives and negatives can also be caused by research bias. Remain objective is vital in eliminating investigator bias.

Statistical and Practical Significance

Statistical significance represents the likelihood that the change in conversion rates between the baseline and the selected variation is not caused by random chance. A statistical result is significant if the value obtained indicates a high likelihood of not occurring by choice (Emmert-Streib, & Dehmer, 2019). Statistical significance evaluates the sample, and the strength of the evidence presented by the results form the basis of whether the result is significant or not p value< 0.05 indicates strong statistical significance.

  To determine statistical significance, the Anova test, P-value, T-tests, and regression coefficients are used. On the other hand, practical significance is used to assess whether the effect displayed by the sample size is significant enough to the real world. Practical significance helps establish the effect of small sample size in the real world (Crowder, 2017). Confidence intervals displaying the upper and lower bounds are used to establish the practical significance behind specific research effectively. Confidence intervals introduce the margin of error to help define the acceptable ranges based on the estimated effect.

Elements of Hypothesis tests

A strong hypothesis is fundamentally the foundation behind strong data-driven optimization. Essentially, formulating a hypothesis is critical in helping break down a wealth of data into meaningful insights. To optimize data- break down, various elements of the Hypothesis have to be tested with a decision made on whether to accept or reject (Crowder, 2017). The hypotheses summarize the assertion about a particular parameter of the population. The Null Hypothesis (H_o) represents a specified population parameter that is considered valid unless concrete evidence suggests otherwise. The Alternative Hypothesis (H₁₎ is formulated to contradict the null Hypothesis.

The alternative Hypothesis is only accepted to be true if the evidence put forward strongly favors it. Hypothesis testing comprises various statistical formulas used to select the null and alternate Hypothesis's feasible option (Emmert-Streib, & Dehmer, 2019). Hypothesis testing has two outcomes. First is the rebuff of the Null Hypothesis and accepting the alternate Hypothesis (H₁). Secondly is the failure to reject the null, therefore, the failure in accepting the alternate.  A P- value </ 0.05 gives a strong indication of concrete evidence against the null. Consequently, the null is retained with the alternate Hypothesis rejected (Emmert-Streib, & Dehmer, 2019). Typically, the null Hypothesis represents the status quo or previously held beliefs.

Drawing a conclusion is also another key element within all hypothesis tests, which plays a significant role in describing how the hypothesis test cannot be used as absolute proof of a null hypothesis. Therefore, failing to provide evidence against the null Hypothesis would mean failing to reject the null Hypothesis. Finding a piece of strong evidence against the null Hypothesis would mean that one will have to reject the null Hypothesis.  The conclusions will then have to be translated into a statement concerning the already existing alternative Hypothesis. Whenever presenting the results of a hypothesis test, it is always advisable to include the descriptive statistics too within the conclusion. The exact p-values should be presented instead of including a specific range.  

Hypothesis Testing

• While testing the Hypothesis, a type 1 error occurs when a false positive result is obtained
• Errors are common while conducting data analysis; this makes it difficult for investigators to avoid type 1 and Type 2 errors completely

Statistical and Practical Significance

• Statistical significance represents the likelihood that the change in conversion rates between the baseline and the selected variation is not caused by random chance
• To determine statistical significance, the Anova test, P-value, T-tests, and regression coefficients are used

Elements of Hypothesis tests

• A strong hypothesis is fundamentally the foundation behind strong data-driven optimization
• The alternative Hypothesis is only accepted to be true if the evidence put forward strongly favors it