Statistical methods used to test the null hypothesis are commonly called tests of statistical significance. What happens is that researchers have the propensity to make statements in this regard: “The difference between the experimental and control group was significant at the .05 level, or the correlation between the two variables was significant at the .05 level” (Gall, 2001). This ideology misinforms statistical researchers and the general public into believing “that the research results are important for this reason” (Gall, 2001). In reality statistical researchers and persons interested in research might be persuaded into surmising that the results of research is focal because it is “statistically significant” (Gall, 2001) or opposite: “that a research result is not important because it is not statistically significant”(Gall, 2001).
The significance tells us that the null hypothesis was redundant due to the “level of certainty” and that accurate circumstances, for example, “random sampling from a defined population”(Gall, 2001) have been satisfied. By accepting the alternative, “the difference between the experimental group and the control groups” (Gall, 2001) is not a result of “sampling error”(Gall, 2001) but that the “samples come from different populations having different mean scores”(Gall, 2001).
Practical significance entails that the “problem warrants some action to be taken to alleviate it” (Choosing a Significance Test, 2006). For example, “if the difference in the incidence of measles between 500 vaccinated and 500 non-vaccinated children is 35%, then vaccination ought to be promoted more actively. However, it the difference is only 2%, it may not warrant any additional action” (Choosing a Significance Test, 2006).
Practical significance is a common misnomer “that only quantitative data obtained through randomization is truly credible and usable” (Practical Significance, 2006). “While certainly data obtained through randomly selected samples is statistically more powerful because of its enhanced ability to reflect the larger campus population, in terms of practical value, non-randomly obtained data (quantitative or qualitative) can be equally as powerful” (Practical Significance, 2006). “Though not as generalizable, this type of quantitative or qualitative data can still yield important and applicable findings on campus climate, serving as an impetus for action planning as well as indicators for change through which progress can be measured over time” (Practical Significance, 2006).
References
Choosing a Significance Test (2006) Steps in data analysis and report writing. Retrieved June 8, 2007 from http://www.idrc.ca/fr/ev-56462-201-1-DO_TOPIC.html
Gall M. D. (2001) Figuring out the importance of research results: Statistical Significance versus Practical Significance. Retrieved June 8, 2007 from http://www.uoregon.edu/~mgall/statistical_significance_v.htm
Practical Significance (2007) Retrieved June 8, 2007 from http://www.oln.org/ILT/ada/Fame/ca/f5_63_529.html
