Factors associated with scientific misconduct and questionable research practices in health professions education

Lauren Maggio; Ting Dong Dong; Erik Driessen; Anthony Artino Jr.

doi:10.1007/S40037-019-0501-X

Figures & Tables

Table 1

Descriptive statistics of ratio scale variables

Variable	Mean	SD	Range
Weighted misconduct frequency score	0.84	0.83	0–4.76
Age	46.03	11.64	23–87
Publication pressure composite score	2.97	0.79	1–5
Number of publications	40.08	54.98	0–600
Percentage of work time doing health professions or medical education	27.32	23.69	0–100
Years involved in health professions or medical education	14.91	9.67	0–54

Table 2

Descriptive statistics of the nominal scale variables

Variable	Category	Frequency	Valid percentage
Gender	Female	305	55.4
Gender	Male	246	44.6
Region of work	North America	246	45.1
	Europe	137	25.1
	Australia/New Zeeland	42	7.7
	Asia	37	6.8
	Africa	53	9.7
	South America	18	3.3
	Others	12	2.2
Academic rank	Trainee	89	16.3
	Junior faculty	163	29.9
	Senior faculty	255	46.8
	Other	38	7.0
Type of research	Quantitative	149	27.2
	Qualitative	119	21.7
	Mixed methods	280	51.1
Work role	Clinician	136	24.7
	Researcher	174	31.6
	Administrator/program director, teacher, or others	240	43.6

Results of two-way Pearson correlation analysis for the variables on a ratio scale are displayed in Tab. 3. As indicated in the table, publication pressure (r = 0.35, p < 0.01) and percentage of work time doing HPE research (r = 0.10, p < 0.05) were both positively correlated with the misconduct score. On the other hand, age (r = −0.19, p < 0.01) and years involved in HPE (r = -0.12, p < 0.01) were both negatively correlated with the misconduct score.

Table 3

Pearson correlation coefficients of all variables except dummy coded nominal variables (region of work, academic rank, type of research, and work role) (N = 590)

	Age	Gender	Years involved in health professions or medical education	Number of publications	Percentage of work time doing health professions or medical education	Publication pressure composite score†	Misconduct score
Age	–	−0.09*	0.81**	0.48**	−0.09*	−0.34**	−0.19**
Gender (male = 1; female = 2)		–	−0.10*	−0.20**	0.09*	0.01	−0.10*
Years involved in health professions or medical education			–	0.50**	−0.05	−0.30**	−0.12*
Number of publications				–	0.10*	−0.24**	0.05
Percentage of work time doing health professions or medical education					–	0.14**	0.10*
Publication pressure composite score						–	0.35**

^†The publication pressure composite score was calculated as an unweighted, mean score for the nine items that comprised the publication pressure scale (Cronbach’s alpha = 0.83)

*P < 0.05; **P < 0.01

Table 4

Three-step hierarchical multiple linear regression modeling. The outcome variable is the square root of weighted misconduct score (N = 590)

Explanatory variable	Block 1			Block 2			Block 3
Explanatory variable	Unstandardized regression coefficient	Standardized regression coefficient	t-test statistic	Unstandardized regression coefficient	Standardized regression coefficient	t-test statistic	Unstandardized regression coefficient	Standardized regression coefficient	t-test statistic
*Block 1*
Age	−0.01	−0.28	−3.86**	−0.01	−0.31	−3.95**	−0.01	−0.22	−2.91**
Gender (Male = 1; Female = 2)	−0.07	−0.07	−1.72	−0.06	−0.06	−1.33	−0.04	−0.05	−1.12
Years involved in health professions or medical education	0.003	0.07	1.03	0.002	0.05	0.68	0.003	0.06	0.84
*Block 2*
Number of publications				0.001	0.13	2.37*	0.001	0.17	3.27**
Region of work (North America as reference group^ξ)
Europe				0.05	0.05	1.05	0.03	0.02	0.54
Australia/NZ				−0.05	−0.03	−0.68	−0.12	−0.07	−1.60
Asia				0.28	0.16	3.53**	0.21	0.12	2.84**
Africa				0.08	0.05	1.19	0.04	0.03	0.57
South America				0.11	0.04	0.93	0.10	0.04	0.93
Academic Rank (junior faculty as reference group)
Trainee				0.005	0.004	0.08	0.04	0.03	0.65
Senior faculty				−0.005	−0.01	−0.10	−0.004	−0.01	−0.09
Type of research (quantitative methods as reference group*)
Qualitative				0.04	0.03	0.63	0.06	0.06	1.15
Mixed-methods				−0.03	−0.03	−0.67	−0.02	−0.03	−0.56
Work role (clinician as reference group)
Researcher				0.11	0.11	1.84	0.12	0.13	2.15*
Other				0.02	0.02	0.36	0.03	0.03	0.60
Percentage of work time doing health professions or medical education				0.0004	0.02	0.40	−0.001	−0.03	−0.72
*Block 3*
Publication pressure composite score							0.20	0.34	7.82**
*Model summary statistics*
R ² change for Block	0.052**			0.058**			0.097**
Total R ²	0.052			0.109			0.206

*P < 0.05; **P <0.01

ξThese groups were selected as the reference groups because they represented the largest respondent groups in their respective categories.