Prof Xavier - Paper and Slides

Reflection – Research Hypothesis

QNT 351

10/23/14

The steps in testing a research hypothesis

There are various steps that are involved in testing a research hypothesis. The first step is usually to make assumptions about the research one is going to conduct. Making assumptions is usually very important in the process of testing a research hypothesis. This is because once you make an assumption, it is very easy to derive hypotheses whether null or true hypothesis. Without the assumptions, it is very difficult to test the hypothesis in a research appropriately. The assumptions are usually made concerning the level of measurement of the variable (Martin, & Bridgmon, 2012).

Secondly, statement of the research and the null hypothesis is what follows the making of the assumptions. In this step, the research hypothesis is usually the one that is in support of the research while the null hypothesis is the hypothesis that is usually in contrast to the topic of research. It is usually stated in negative terms. It is always null unless tested otherwise at the end of the research.

Selection of the distribution of sampling is what follows. It is important to select the sampling distribution because the hypothesis has to be applied on some statistics for them to be tested (Martin, & Bridgmon, 2012).

After the selection, computation of the statistics in order to test the hypothesis deduced is what follows. In this step, the sampling distribution is used. Finally, the decision is made on whether the hypothesis was true of false. In this step, interpretations of the sampling distribution are done.

Comparing the means of two or more groups

There are two steps involved in the process of comparing the means of two or more groups. To begin with, the comparison means that there must have been two or more groups that were selected as the samples for distribution. This could have been may be a control group and the actual group.

The first step is to carry out an ANOVA analysis. This is referred to as the analysis of variance. Once the variance is analyzed, it is possible to tell whether all the means to the groups are equal or whether they vary and if they do, by how much (Koop, 2000).

Secondly, if the means of the groups are different, the test is done to show the significance of the difference that exists between the means of the groups. In comparing the means of the two groups, the honest significance difference is found to show the spread of the means of the two groups.

Calculating the correlation between two variables

In calculating the correlation between two variables, the first step is to label the variables as either independent on the x axis and the depended on the y-axis. After labelling the variables, one creates a table on which the values of the two variables are filled (Koop, 2000).

The total values of each variable are taken for instance; values for x are filled in column A while those of Y are filled in column C. The squares of X and Y are filled in column B and D consecutively. Once the totals have been arrived at, computation is done to arrive at the correlation between the two variables.

The correlation prediction of whether a change in one variable is likely to affect the other variable to the extent of producing a proportional change on the variable.

References

Koop, G. (2000). Analysis of economic data. Chichester: John Wiley.

Martin, W. E., & Bridgmon, K. D. (2012). Quantitative and Statistical Research Methods:

From Hypothesis to Results. Hoboken: John