Statistical tools and analysis provide researchers some data to confirm or reject hypothesized assumptions. One of the most commonly utilized analysis to identify a number of results in a group or comparing parts of two or more groups is Proportions. But what is proportions analysis? Proportions analysis in statistics intends to come up with a good perspective whether a part of the data set results follows a useful concept for making decisions or confirming an assumption.
In layman’s term, a proportion is finding the representative data part compared to the whole of the data set, used to compare two ratios (Algebra Help, 2001). Using proportions analysis in the study for Alternative Educational Course for ADHD is somehow significant and efficient. Basically, it is possible to conduct surveys and experimental analyses that will come up with the binary results of responses. For example, it is possible to tally the ‘good’ effect and the ‘bad’ effects of the Alternative ADHD course which simply limits to two (binary) answers.Therefore, we can proportionate the results coming from the good responses and from the bad responses.
There are two possible scenarios that can be applied in using proportions for the ADHD study dataset. One is by analyzing proportions within a single population or group and another by analyzing the comparative proportions between two groups. In the first scenario, doing an analysis in a single group of responses can be achieved. For example, if we are to assume that 10% or fathers (male parents) think that the Alternative ADHD is successful, then we can use proportions analysis to confirm this.
The details that are needed are the actual number of males who said they think the Alternative study was successful, the total number of male parents and the choice of level of significance. Upon getting all these information, a One Sample Z-test for the proportion can be done to see whether we Reject the hypothesis (10% of the male respondents think the ADHD Alternative study was successful) or not. A One Sample Z-test is used to see whether a randomly selected sample data set have the same characteristics of its population (Majesty).Apart from using the proportions analysis for a single group, the proportions analysis can also be conducted to compare two groups of the dataset respondents. This is called the Two Sample Z-test for difference in two proportions. The main goal is to see whether there is a significant difference between two proportions of two categories of the data set (Stat Trek, 2007).
For example, we would like to analyze whether the proportions of males who think the ADHD Alternative study was successful is different from the proportions of females who also think the same.The details that are needed for such an analysis are the level of significance, the numerical hypothesized difference, the number of males who said ‘yes’ and the total number of males and the number of females who said ‘yes’ and the total number of females. It does not really matter whether the total number of males and total number of females are different because the proportions in each group will make up for such a concern.By doing the Z-test for the difference of two proportions, we can conclude whether to accept or reject the hypothesis that there is a difference of responses between the two gender groups. The presented Z-test for two populations can only be used if the following criterions are met; the sampling method used was Simple Random Sampling, the responses from both groups are independent, each sample has at least 10 successes and failures and each population is at least 10x as large is its samples (Stat Trek, 2007).
