Chanakya Research has emerged as one of the most successful firms catering high end statistical data analysis support for researchers struggling with the data analysis part of their thesis/dissertations. All our statisticians possess a brilliant amount of expertise in using the finest statistical testing tools which can ensure the best results for your research task. We use a holistic approach to ensure that the data collected as part of the research assignment is included within the thesis/dissertation in an accurate and valid format. Prior to the usage of a specific statistical tool, we ensure to go through the actual aim behind the research conducted by our client.
The question that you encounter while doing your analysis is: “How well does one variable relates to another variable under study?”. Although, through statistical analysis, you get the answer to this question by testing the probability of occurrence of the relationship between the variables if there existed no difference in the population from which the sample was drawn. This process is known as significance or hypothesis testing, as in effect you are comparing the data you have collected with what you would theoretically expect to happen. There are two main groups of statistical significance tests: non-parametric and parametric. Non-parametric statistics are designed to be used when your data are not normally distributed. Not surprisingly, this most often means they are used with categorical data. In contrast, parametric statistics are used with numerical data.
Testing the probability of a pattern such as a relationship between variables occurring by chance alone is known as significance testing. While carrying out your research, you have to examine the collected data from the samples to analyse the relationship among variables. Once you enter data into the analysis software, choose the statistic and click on the appropriate icon, the answer appears as if by magic! Most of the statistical analysis software are based on the test statistic, the degrees of freedom (df) based on which the probability value of your test is ascertained. If the probability of your test statistic of one or more extreme having occurred by chance is very low (say, p< 0.05 or lower), then you have a statistically significant relationship. Statisticians refer to this as rejecting the null hypothesis and accepting the hypothesis, often abbreviating the terms null hypothesis to Ho and Hypothesis to H1. Consequently, rejecting a null hypothesis will mean rejecting a testable statement something like ‘there is no significant difference between…’ and accepting a testable statement something like ‘there is a significant between…’. If the probability of obtaining the relationship between variables under test is higher than 0.05, then it is concluded that there is no statistically significant relationship. Statisticians refer to this as accepting the null hypothesis. There may still be a relationship between the variables under such circumstances, but you cannot make the conclusion with any certainty.
Often descriptive or numerical data is summarised as a two-way likelihood table. It is based on a comparison of the observed values in the table with what might be expected if the two distributions were entirely independent. The test relies on:
- The mutually exclusive categories being used in the table, so that each observation falls under only one category or class interval;
- No more than 25 per cent of the cells in the table having expected values of less than 5. For contingency tables of two rows and two columns, no expected values of less than 10 are preferable.
If the latter assumption is not met, the accepted solution is to combine rows and columns where this produces meaningful data.
Chanakya Research provides full-fledged statistical data analysis support for researchers , and always up to respond to your query. Contact us at firstname.lastname@example.org to get assistance in the best possible manner.