Hypothesis Tests

HypothesisTests

HypothesisTest

A theory testhelps evaluate the association between two statements that areseparate and identify one, which is best supported by the availabledata (Baltaxe, Meer &amp Lindenbaum, 2016). The paper will discuss aworking example of a hypothesis test.

For example, aconsumer watch group want to determine the changes in electricitybill from the past years when the average bill per household was $240. Consequently, the group samples 25 families for the currentelectricity bills.

The results ofthe study are, the current mean cost, $ 330, the standard deviationfor the data is 149.8. A look at the data shows a $70 increase in theelectricity bills. However, this data does not represent the wholetruth about the situation. The average of the random sample (25households) is $330, but it is likely that the mean of the populationis close to the $260 calculated in the previous studies. In otherwords, errors occur in sampling, with repeated sampling. The errorscan be minimized by multiple sampling of the particular population.The average would still range around $ 260 (Baltaxe, Meer &ampLindenbaum, 2016). Hypothesis tests help in evaluating such apossibility.

In almost allcases the sample mean does not equal the population average. One wayof reducing the difference is by taking multiple samples of thepopulation and representing the sample statistic (in this case mean)on a sampling distribution. However, this method consumes time andenergy. A t-test can be used to determine the variability in theaverage by calculating the standard deviation (Baltaxe, Meer &ampLindenbaum, 2016).

In conclusion,the distribution curve above shows the mean average for thepopulation is likely to be around $ 260. However, there is thepossibility to get a higher or lower average. Therefore, a hypothesistest helps in finding the statistical significance of the deviationbetween two sets of data.

Reference

Baltaxe, M., Meer, P., &amp Lindenbaum, M. (2016). LocalVariation as a Statistical Hypothesis Test.&nbspInternationalJournal of Computer Vision,&nbsp117(2), 131-141.