EssayExperts.ca - Premium Custom Essays 1.877.974.8398

Toronto: 416.222.5225         Vancouver: 604.757.5246
                    support@essayexperts.ca

Essay Writing

We've successfully helped 50,000+ university students in Canada with our premium custom essays.

Thesis & Dissertation

Our Thesis and Dissertation division has helped over 15,000 graduate students whether it's one chapter or all five. Find out more.

How We Work

Quick Essay Service Links
Have a Custom Request?

We've written unique requests from books, bios, grants and more. Order any project today.

Order Custom Request

31 Essay Tips - Sign Up

31 Essay Tip Course

How to Order

Current Service Hours

Mon-Wed:  10 am - 11 pm
Thu-Fri:  10 am - 11 pm
Sat-Sun:    1 pm - 9 pm
------------------

*Walk-in Visit by Appointment Only

Join Us Today
Expert Extras
Spain Flag Korea Flag France Flag Russia Flag China Flag Iran Flag
EssayExperts

Tests of Statistical Significance

Understanding tests of statistical significance (relationship between two variables)

statistical significance

When conducting different statistical test such as chi-square test of association, t-tests, ANOVA, and many others, tests of statistical significance are used to determine the probability that an observed relationship between two measures is entirely by chance. For example, if we took 4 random samples from a population, would it be possible to still find the same relationship between two variables in every sample that has been randomly selected? Furthermore, if we could get the data for the entire population, would it be possible to find the same relationship, or would the findings from the sample be an occurrence only by chance. Key highlights of test of statistical significance;

  • All tests of statistical significance usually help a researcher determine the probability that they would be making a mistake if they assumed that a certain relationship exists between two variables.
  • Because one cannot be absolutely 100% that a certain relationship between two variables. We use tests of statistical significance to help understand the probability of being wrong. Uncertainty in statistical analysis comes from, sampling errors, study bias, data reliability and other simple mistakes that cannot be measured.
  • When working with a test of statistical significance, one basically estimates the probability that they might be wrong after making the assumptions that they are right (assumed a certain relationship exists). In this case, if the computed probability of being wrong is significantly small, a researcher can conclude that their finding about the relationship is statistically significant.
  • In simpler words, statistical significance is the assumptions that the observed finding is correct and that it does exists. However, it should not be confused with practical implications. A statistically Significant finding does snot necessarily have to be practically valid in real life.
Errors in tests of statistical significance

statistics tests

There is usually a chance that a researcher will end up making mistake in regard to their findings or assumptions about a certain relationship between two variables when using a test of statistical significance to guide their study. The two common mistakes made are either a type I error or a type II error.

Type I error

The type I error also known as (false positive) is committed when the researcher rejects a null hypothesis when the null hypothesis is in fact true. In simpler words, based on a test of relationship, a type I error occurs when a researcher concludes that a relationship does exist between two variables when in fact, the actual evidence is that the relationship does not exist.

Type II error

The type II error also known as (false negative) is committed when the researcher accepts a null hypothesis when the null hypothesis should have been rejected. In simpler words, based on a test of relationship, a type II error occurs when a researcher concludes that a relationship does not exist between two variables when in fact, the actual evidence is that the relationship does exist.

Key highlights of type I and type II errors

In cluster sampling the population is segmented into different subgroups. Each of these subgroups shares the same characteristics as the rest of the subgroups and in clustered sampling, one chooses entire subgroups at random, rather than individual members from the population.

  • Various measures are taken to reduce the chance of committing the errors, however, when trying to reduce the chance of committing a type I error a researcher ends up increasing the chance of committing a type II error and the vice versa is also true.
  • In many studies, the researcher will try and reduce the chances of committing a type I error. This is because, when one commits a type I error, the conclusion is that a relationship exists when it does not which is worse than a type II error which assumes that the relationship does not exist therefore, not changing a lot.

Essay Experts is Canada's premier essay writing and research service. We help undergraduate and graduate students with their essays, research papers, theses and dissertations. Our statisticians are standing by to help. Simply email us your question, requirements or assignment and we'll get back to you with a quote. Our statisticians all possess advanced degrees and have experience in helping students succeeed in statistical writing and analysis.

Order Now!