Skip to main content

Table 1 Articles that misused ANOVA from the current issues of health-related journals published by the African Journals Online

From: Misuse of analysis of variance in African biomedical journals: a call for more vigilance

S. No.

Journal

Dependent variable type

Amount of categorical independent variable

Independence of observation

Post hoc test

ANOVA misuse

References

1

Afri. J. Clin. Exp. Microbiol

Continuous

6

Yes

No

Did not perform ANOVA

Adeyemi et al. (2020)

2

Afri. J. Reprod. Health

Continuous

3

Yes

Yes

One-way ANOVA

Lee et al. (2020)

3

Afri. J. Complement Altern Med

Continuous

4

Yes

No

Used t test instead of one-way ANOVA

Iikasha et al. (2019)

4

Afri. J. Urology

Continuous

3

Yes

No

Authors did not mention the type of ANOVA used

Ahmad et al. (2018)

5

Afri. J. Urology

Continuous

2

Yes

Yes

Authors did not mention the type of ANOVA used

Ali et al. (2018)

6

Alexandria J. Med

Continuous

3

Yes

Yes

Authors did not mention the type of ANOVA used

Assaad et al. (2018)

7

Alexandria J. Med

Continuous

4

Yes

Yes

Authors did not mention the type of ANOVA used

Korany et al. (2018)

8

Alexandria J. Med

Continuous

5

Yes

No

The authors did not mention the type of ANOVA used

Emara et al. (2018)

9

Alexandria J. Med

Continuous

4

Yes

Yes

Did not specify the type of ANOVA used

Khowailed et al. (2018)

10

Alexandria J. Med

Continuous

4

Yes

No

The authors did not mention the type of ANOVA they used and did not use means for ANOVA

Ephraim et al. (2018)

11

Alexandria J. Med

Continuous

4

Yes

No

The data showed sig among groups without a post hoc analysis

Sherbiny et al. (2018)

12

Alexandria J. Med

Continuous

2

Yes

No

Mentioned ANOVA though did not perform ANOVA and was not supposed to perform ANOVA

Igbigbi et al. (2018)

13

Anat. J. Afri

Continuous

2

Yes

No

Mentioned ANOVA though did not perform ANOVA and was not supposed to perform ANOVA

Idenya and Gichangi (2020)

14

Arch. Med. Biomed. Res

Continuous

3

Yes

No

Mentioned ANOVA test but authors performed chi-square test

David et al. (2018)

15

Arch. Med. Biomed. Res

Continuous

4

Yes

Yes

Authors did not mention the type of ANOVA used

Akindele et al. (2018)

16

Egyptian J. Pediatr. Allergy Immunol

Continuous

3

Yes

No

Mentioned ANOVA though did not perform ANOVA and was not supposed to perform ANOVA

Osman et al. (2019)

17

Ethiopian J. Heal. Dev

Continuous

4

Yes

Presumably yes. But the authors did not mention the post hoc

The type of ANOVA was mentioned

Han et al. (2020)

18

IJBAIR

Continuous

1

Not available

No

Wrong mention of ANOVA. Authors did not perform ANOVA

Muhammad et al. (2018)

19

IJBAIR

Continuous

5

Yes

No

Means were not used for the one-way ANOVA

Oise et al. (2018)

20

IJHPR

Continuous

4

Yes

Yes, presumably. But the authors did not mention the post hoc

The type of ANOVA test was not mentioned

Iyomon et al. (2016)

21

J. Med. Biomed. Res

Continuous

8

 

No

The type of ANOVA test was not mentioned

Ehwarieme and Anarado (2017)

22

Nig. J. Physiol. Sci

Continuous

3

Yes

No

Authors mentioned ANOVA but used t test

Ogunro et al. (2019)

23

Nig. J. Physiol. Sci

Continuous

4

Yes

Yes

Did not mention the type of ANOVA

Kunle-Alabi et al. (2019)

24

Nig. J. Physiol. Sci

Continuous

6

Yes

Yes

Mentioned the use of two-way ANOVA, while the authors used one-way ANOVA

Adeneye et al. (2019)

25

Res. J. Heal. Sci

Continuous

4

Yes

Yes

Did not mention the type of ANOVA

Sanni et al. (2020)

26

Res. J. Heal. Sci

Continuous

5

Yes

No

Did not perform ANOVA correctly. The means of each group was not calculated

Uduagbamen et al. (2020)

27

TJPR

Continuous

2

Yes

No

Did not perform ANOVA and was not supposed to do so but the authors mentioned it

Yuan et al. (2020)

  1. Checklist for correct use of ANOVA statistical test based on which all the articles assessed
  2. That each sample has been drawn independent from one another. This is usually verified in the method section of a research article
  3. That the dependent variable is continous
  4. For specific cases where the authors described the significant difference of a group from others in the discussion, they ought to have performed a post hoc test after the ANOVA
  5. That the type of ANOVA used is mentioned at least once in the article
  6. That the data sets analysed are means comprising of at least three scores per variable
  7. That the data set conforms with the type of ANOVA employed, e.g. mentioning one-way ANOVA where a data set having two categorical independent variable will be a misuse