Statistical Significance and Meaningfulness
Please let me know if this homework disucssion requires more than one page. The below resources can be found under the Manage Orders section:
- Instructions with hyperlinks to some resources
- Meaningfulness vs. Stat. Signifigance
- ASA – American Stat. Association
- Applied statistics from bivariate through multivariate techniques (2nd ed.)
- Skill builders (if needed)
Discussion: statistical significance and meaningfulness
The purpose of this assignment is to discuss statistical significance and meaningfulness with reference to the statement and footnote below.
“A research paper claims a meaningful contribution to the literature based on finding statistically significant relationships between predictor and response variables. In the footnotes, you see the following statement, “given this research was exploratory in nature, traditional levels of significance to reject the null hypotheses were relaxed to the .10 level.”
Hypothesis testing, meaningfulness, and statistical significance
Research involves contribution to knowledge by using data to infer meaning and using the calculate statistics to generalize results over a wide population. Key among the output of a results is information about the predictor and response variables. Statistically significant results show existence of relationships between the predictor and response variables. Using a null hypothesis the decision to reject is reached if the p-value is lesser than a stated alpha value which is 1 minus the confidence level. The logic behind Null hypothesis significance testing (NHST) involves a ‘guess’ about the specific value of a parameter for a population of interest. The researcher then calculates a critical value form the sample and compare it with the population parameter and if the value is far from the hypothesized population parameter giving a huge z ratio thus the probability of finding a similar value from a population is large.
NHST is problematic in that people have a strong tendency to state hypothesis that they believe are correct and then look for evidence to confirm their hypothesis. Secondly the logic assumes a random sample which is not always the case as researchers use convenient sample which makes it hard to generalize the data. Using the NHST logic presents a likelihood of making either type I or type II errors. Type II error occur when a researcher fails to reject a null hypothesis when it is actually false while Type I error occur when a researcher reject a null hypothesis when it is actually true (Frankfort-Nachmias, & Leon-Guerrero, 2015; Wagner, 2016).
However, the use of p-value significance has faced a lot of criticism from different researchers. According to some of the researchers, good statistical practice is a key component to realizing a good scientific practice. In order to come up with good results a good study design and conduct is very critical. This data should also be summarized properly and interpretation done within the required context to enable understanding of what the summaries mean (Wagner, 2016; Warner, 2012). P-value is one of the key statistic used in determining significance and was never intended to substitute scientific reasoning but a statistic to bolster the reasoned results. According to research p-values can be significant even with incompatible data thus resulting to wrong inferences. In addition p-value do not measure the probability that a hypothesis set is true and thus even unrepresentative data may produce significant p-values. Moreover p-value does not measure the size of an effect of one variable towards another (Frankfort-Nachmias, & Leon-Guerrero, 2015; Warner, 2012).
Frankfort-Nachmias, C., & Leon-Guerrero, A. (2015). Social statistics for a diverse society (7th ed.). Thousand Oaks, CA: Sage Publications. Chapter 9, “Testing Hypothesis” (pp. 267– 277)
Wagner, W. E. (2016). Using IBM® SPSS® statistics for research methods and social science statistics (6th ed.). Thousand Oaks, CA: Sage Publications. Chapter 6, “Testing Hypotheses Using Means and Cross-Tabulation”
Warner, R. M. (2012). Applied statistics from bivariate through multivariate techniques (2nd ed.). Thousand Oaks, CA: Sage Publications. Chapter 3, “Statistical Significance Testing” (pp. 81–124)