Statistical significance refers to a statistic being reliable; it does not mean that the data gathered would certainly have any decision making utility. It allows the researcher to compute the level of strength of the relationship between two or more variables. For instance, the extent of the significance of a relationship can be evaluated using statistical significance analysis. The statistical tests help provide the “p-value” that represents the probability of a random chance to explain the findings. In most cases, values less than 5% are considered to be statistically significant.
The concept of statistical significance can be understood using the following example; a company wants to formulate a production plan for its shoes. It wants to determine the various sizes to make available for its men and women’s shoes. The company does not want to base its production base on the anecdotal fact that men usually have bigger feet than women; it may conduct a statistical significance test to identify the degree of correlation between genders and foot size and set its production base accordingly.
If the research shows for instance, a 3% p- value, then the company can conclude that the results are statistically significant. It can base its production plan on the data gathered from the study. The 3% value indicates that the relationship between gender and foot size was only a result of chance. If the value is greater than 5%, then it may suggest otherwise.
One of the drawbacks of using statistical analysis is that the data may be statistically significant but not meaningful; this implies that the relationship can be deemed as having statistical significance if it can be proved using numbers and statistics, but it may have no real life value or meaning.
In cases where the design of the study is poorly constructed, the results of the research would not be meaningful. Statistical significance is not analogous to practical importance of the study. For instance, if a study suggests difference between two types of population that is not due to chance alone, it does not show what caused the difference. There is a majority of intervening and extraneous factors that play an important role in influencing cause-effect relationships therefore; the research must be carefully designed. The meaningfulness of a research lies in the qualitative aspect of the study. For example; 10% heart surgeries fail on an average basis, the study employed a 5% p-value, hence for a statistician the data would not be statistically significant but it does carry meaningful significance for those people who are considering getting the surgery done.
A lot of researchers use the word, ‘significant’ to describe the findings of the study as carrying decision making utility. However, this is an incorrect use of the word from a statistician’s viewpoint. The word is understood universally as having the same meaning as important/noteworthy. Thus, many researchers use the word ‘significant’ to explain the relationship between two or more variables that may be strategically important to a client, in spite of any statistical tests. The client is, therefore, advised to be mindful about the relationship as it would be of great importance to the organization’s strategic plan. The statisticians must also be careful when using the word ‘significance’; they should use the complete term, ‘statistical significance’ to refer to the statistical relationship between variables when communicating with the general public to avoid misinterpretation and confusion.
Free Research Essay Sample
Type of paper: Essay
Topic: Education, Relationships, Company, Planning, Utilitarianism, Women, Study, Information
Pages: 2
Words: 600
Published: 03/31/2020
Cite this page
- APA
- MLA
- Harvard
- Vancouver
- Chicago
- ASA
- IEEE
- AMA