Skip to main content

Table 2 Estimates and model fit from the latent growth curve model

From: A modest start, but a steady rise in research use: a longitudinal study of nurses during the first five years in professional life

  

Evaluation of model fit

Longitudinal main effects

Variability

Estimator

 

χ 2§

CFI

RMSEA (C)

SRMR

Intercept

Slope

Cov (I, S)

Var (I)

Var (S)

Instrumental research use

ML a

47.3***

0.95

0.05 (.48)

0.06

2.84***

0.19***

-0.09**

0.99***

0.06***

 

ML a, b

46.2***

0.95

0.05 (.37)

0.06

2.86***

0.14**

-0.09**

1.00***

0.06***

 

MLR a

44.7***

0.94

0.05 (.56)

0.06

2.84***

0.19***

-0.09**

0.99***

0.06***

 

WLS a

97.2***

0.95

0.06 (.16)

-

0

0.14**

-0.04**

0.52***

0.03***

 

ML c

43.4***

0.92

0.08 (.02)

0.08

2.92***

0.18***

-0.07 ns

0.95***

0.05**

 

ML a, d

24.4**

0.98

0.03 (.95)

0.05

2.91***

0.17***

-0.05**

0.95***

0.04***

Conceptual research use

ML a

110.1***

0.72

0.08 (.01)

0.06

2.58***

0.25***

-0.07**

0.61***

0.05***

 

ML a, b

109.2***

0.72

0.09 (.01)

0.06

2.57***

0.29***

-0.07**

0.60***

0.05***

 

MLR a

105.4***

0.71

0.08 (.01)

0.06

2.58***

0.25***

-0.07**

0.61***

0.05***

 

WLS a

209.0***

0.67

0.09 (.01)

-

0

0.21***

-0.05**

0.41***

0.04***

 

ML c

70.7***

0.76

0.11 (.01)

0.08

2.58***

0.26***

-0.09*

0.69***

0.06***

 

ML a, d

41.1***

0.91

0.05 (.52)

0.04

2.71***

0.23***

-0.03 ns

0.52***

0.03***

Persuasive research use

ML a

31.2***

0.96

0.04 (.96)

0.04

1.71***

0.09***

0.01 ns

0.29***

0.02**

 

ML a, b

31.0***

0.96

0.04 (.83)

0.04

1.71***

0.07*

0.01 ns

0.29***

0.02**

 

MLR a

24.2**

0.96

0.03 (.98)

0.04

1.71***

0.09***

0.01 ns

0.29***

0.02**

 

WLS a

39.2**

0.99

0.03 (.99)

-

0

0.10***

-0.02 ns

0.48***

0.02**

 

ML c

14.6 ns

0.99

0.02 (.99)

0.04

1.76***

0.10***

0.01 ns

0.32***

0.02**

 

ML a, d

15.3 ns

0.99

0.02 (.99)

0.03

1.75***

0.08***

0.01 ns

0.29***

0.02**

  1. a Estimation includes all available data from any of the five data collections.
  2. b Model includes an additional quadratic main effect.
  3. c Estimation includes only those with complete RU data from all five data collections.
  4. d Revised model, freeing the second time point from being included in the linear model.
  5. § All models have 10° of freedom except the model with a quadratic effect (df = 9), the WLS estimated model (df = 18), and the revised model (df = 9).
  6. Abbreviations: ML = Maximum Likelihood; MLR = Robust Maximum Likelihood; WLS = Weighted Least Square (note that SRMR are not computed with this estimator and that estimated effects are on another scale); CFI = Comparative Fit Index; RMSEA = Root Mean Square Error of Approximation; (C) = p-value for RMSEA deviating from 0.05; Cov (I, S) = covariance between variability in intercept and slope; Var (I) = intercept variance; Var (S) = slope variance; ns = non significant; *p < .05; **p < .01; ***p < .001.
Back to article page

Implementation Science

ISSN: 1748-5908