Skip to main content
. 2016 May 23;35(2):186–209. doi: 10.1080/14636778.2016.1184965

Table 3. Ordered regression predicting perceptions of bioinformatics as a discipline and a service.

  Model 1: Discipline Model 2: Service
B SE Wald Exp(B) B SE Wald Exp(B)
Dependent variable
 Totally disagree −1.13 .77 2.15   −2.47 .77 10.39  
 Disagree 0.08 .76 0.01   −1.28 .76 2.83  
 Neither agree nor disagree 0.76 .76 0.99   −0.67 .76 0.77  
 Agree 2.64 .78 11.37   0.99 .76 1.69  
Ref: totally agree
Independent variables
Controls
 Gender 0.00 .19 0.00 1.0 0.09 .19 0.24 1.1
 Seniority −0.10*** .04 6.11 0.9 −0.06* .04 2.62 0.9
UG degree period
 During HGP 0.01 .17 0.01 1.0 −0.29** .18 2.77 0.7
 Post-HGP 0.34 .31 1.18 1.4 −0.44* .31 1.99 0.6
Ref: Pre-HGP
Discipline
 Bioinformatics 0.07 .17 0.16 1.1 −0.50*** .18 8.01 0.6
 Biology −0.45*** .15 8.54 0.6 0.05 .15 0.09 1.0
 Medicine −0.01 .26 0.00 1.0 −0.10 .26 0.16 0.9
 Computer Science −0.18 .18 1.01 0.8 0.01 .18 0.00 1.0
 Mathematics −0.05 .25 0.05 0.9 0.41* .25 2.66 1.5
 Statistics −0.26* .19 1.99 0.8 −0.08 .19 0.17 0.9
Funding source
 RCUK −0.32** .17 3.69 0.7 −0.34** .17 4.12 0.7
 Charity 0.09 .16 0.33 1.1 −0.33** .16 3.99 0.7
 NHS −0.07 .34 0.04 0.9 −0.26 .36 0.52 0.8
 Commercial 0.16 .19 0.73 1.2 0.20 .19 1.06 1.2
 EU 0.03 .16 0.03 1.0 −0.25* .16 2.46 0.8
Esteem indicators
 Software 0.12** .06 3.81 1.1 0.09* .06 1.94 1.1
 Funding 0.16** .07 4.95 1.2 −0.11* .07 2.30 0.9
 Teaching 0.22*** .07 9.93 1.2 0.03 .07 0.13 1.0
 Papers −0.10 .09 1.26 0.9 −0.16** .09 2.90 0.9
 Service −0.06 .06 0.90 0.9 0.01 .06 0.01 1.0
 PhD supervision −0.03 .08 0.10 1.0 −0.13* .08 2.45 0.9
 Conference 0.07 .08 0.88 1.1 0.04 .08 0.23 1.0
 Patents 0.00 .09 0.00 1.0 0.09 .09 0.99 1.1
 Commercial 0.03 .08 0.11 1.0 0.05 .08 0.37 1.1
Modes of learning
 Informal −0.01 .11 0.00 1.0 −0.23** .11 4.71 0.8
 Formal 0.02 .07 0.14 1.0 0.03 .06 0.21 1.0
Perceptions
 Imp. bckgrnd medicine 0.03 .09 0.10 1.0 0.48*** .09 26.36 1.6
 Imp. bckgrnd comp sci 0.12* .09 1.85 1.1 0.09 .09 1.06 1.1
Model fit
 −2 log likelihood 598.890     615.910    
 Model χ2 57.381     82.566    
 df 28     28    
 sig. .001     .000    
Na 245     242    
 Nagelkerke Pseudo R2   .22       .31    

Notes: B = coefficient (mean change in the response variable for one unit of change in the predictor variable while holding other predictors in the model constant); SE = standard error (the standard error around the coefficient for the constant); Wald = Wald Test; Exp(B) = the exponentiation of the B coefficient, which is an odds ratio (this value is given by default because odds ratios can be easier to interpret than the coefficient, which is in log-odds units).

aReduction in sample size due to listwise deletion of cases necessary for regression requirements.

*Level of statistical significance: p < .10.

**Level of statistical significance: p < .05.

***Level of statistical significance: p < .01.