1 of 8

Statistical Tests in Science

Amberly Bailey

2 of 8

Standard

LS1.C Organization for Matter and Energy Flow in Organisms

  • Plants need water and light to live and grow

3 of 8

Purpose of Statistical Tests

  • Statistical tests serve to examine a scientific question by comparing two or more groups
  • A statistical test calculates the probability of a hypothesis

​

​

​

​

(Victor et al. 2010)

4 of 8

Steps of Statistical Tests

  1. Statement of the question to be answered by the study
  2. Formulation of hypotheses
  3. Choosing a suitable test
  4. Specify level of significance (for example, 0.05)
  5. Performance of test and calculation of p-value
  6. Statistical decision (example below)
  7. p<0.05 → significant evidence to support hypothesis
  8. P≥0.05 → no significant evidence to support hypothesis
  9. Interpret test result

5 of 8

Common Statistical Tests in Science

  • Unpaired t-Test:

An unpaired t-test is usually used to compare just two sets of data

(Nayak and Hazra, 2011)

​

Height

Shoe Size

5’0

5

5’2

6

5’4

7

5’6

8

6’0

10

6 of 8

Common Statistical Tests in Science

  • ANOVA:

An ANOVA test is used to compare more than two sets of numerical data

If the results show significant evidence (p<0.05), a post-hoc test should be done!

​

(Nayak and Hazra, 2011)

7 of 8

Common Statistical Tests in Science

  • Post Hoc:

Post Hoc tests are used if there is significant evidence in an ANOVA test.

They are used to find certain differences between groups.

​

(Nayak and Hazra, 2011)

Fun Fact: “post hoc” means “after this” in Latin!

​

8 of 8

Literature Cited

Nayak BK, Hazra A. 2011. How to Choose the Right Statistical Test? Indian

Journal of Ophthalmology 59(2):85-86.

​

Victor A, Elsässer A, Hommel G, Blettner M. 2010. Judging a Plethora of p-values:

How to Contend with the Problem of Multiple Testing. Dtsch Arztebl

International 107(4):50–6.

​

​