Search Results

You are looking at 1 - 10 of 11 items for :

  • "imputation" x
  • Athletic Training, Therapy, and Rehabilitation x
Clear All
Restricted access

Brian G. Pietrosimone, Adam S. Lepley, Hayley M. Ericksen, Phillip A. Gribble and Jason Levine

Background:

Disability is common in a proportion of patients after anterior cruciate ligament reconstruction (ACL-R). Neuromuscular quadriceps deficits are a hallmark impairment after ACL-R, yet the link between muscle function and disability is not understood.

Purposes:

To evaluate the ability of quadriceps strength and cortical excitability to predict self-reported disability in patients with ACL-R.

Methods:

Fifteen participants with a history of ACL-R (11 female, 4 male; 172 ± 9.8 cm, 70.4 ± 17.5 kg, 54.4 ± 40.9 mo postsurgery) were included in this study. Corticospinal excitability was assessed using active motor thresholds (AMT), while strength was assessed with maximal voluntary isometric contractions (MVIC). Both voluntary strength and corticospinal excitability were used to predict disability measured with the International Knee Documentation Committee Index (IKDC).

Results:

The overall multiple-regression model significantly predicted 66% of the variance in self-reported disability as measured by the IKDC index (R 2 = .66, P = .01). Initial imputation of MVIC into the model accounted for 61% (R 2 = .61, P = .01) of the variance in IKDC. The subsequent addition of AMT into the model accounted for an insignificant increase of 5% (Δ R 2 = .05, P = .19) in the prediction capability of the model.

Conclusions:

Quadriceps voluntary strength and cortical excitability predicted two-thirds of the variance in disability of patients with ACL-R, with strength accounting for virtually all of the predictive capability of the model.

Restricted access

Frank E. DiLiberto, Deborah A. Nawoczenski and Jeff Houck

maintain the statistical power and a sample size of 12 participants, missing data for the 2 cases pertaining to the high-step activity were addressed via a mean imputation method with random variability that was based on the present distribution. 31 , 32 Mean imputation with random variability reduces

Restricted access

Jesse C. Christensen, Caitlin J. Miller, Ryan D. Burns and Hugh S. West

patient for KOS-ADL (final score−admission score) and NPRS (admission score−final score) outcomes. All patients in the analysis completed the patient-reported outcomes at both time points; therefore, there was no concern for response bias or potential estimate errors based on statistical imputation for

Restricted access

Zachary Y. Kerr, Andrew E. Lincoln, Shane V. Caswell, David A. Klossner, Nina Walker and Thomas P. Dompier

of each category AE count do not equal total AE count due to rounding error (due to the use of mean imputation values based on all other valid AE data from the same year, division, and event type for missing data). Table 2 Injury Frequencies and Rates With 95% Confidence Intervals (CI) by Season and

Restricted access

Jeffrey Martin, Mario Vassallo, Jacklyn Carrico and Ellen Armstrong

imputation for five missing data points on the judge’s ratings of emotion was used. With this very low percentage of missing data, this approach is acceptable over more contemporary approaches such as multiple imputation ( Schafer, 1999 ; Schlomer, Bauman, & Card, 2010 ). We also calculated interrater

Restricted access

Jay Johnson, Michelle D. Guerrero, Margery Holman, Jessica W. Chin and Mary Anne Signer-Kroeker

imputation procedures using the fully conditioned specification approach. A total of five imputed data sets were generated, with each data set comprising slightly different imputed values. The parameter estimates of the five imputations were pooled and yielded a single set of estimates. Given that little is

Restricted access

Chunxiao Li, Ngai Kiu Wong, Raymond K.W. Sum and Chung Wah Yu

through a single imputation using an expectation–maximization algorithm, χ 2 (629) =629.30, p  = .49 ( Little, 1998 ). The means, SD s, internal reliability, and zero-order correlations of the study variables were computed after cleaning the data. All these statistical analyses were performed with IBM

Open access

Terese Wilhelmsen, Marit Sørensen and Ørnulf N. Seippel

platforms for the analyses. To handle missing values in the data, we used the R package “Multivariate Imputation by Chained Equations (MICE)” ( Van Buuren & Groothuis-Oudshoorn, 2011 ). To perform the two-step fsQCA analyses ( Schneider & Wagemann, 2006 ), we used the R package “QCAQUI” ( Dusa, 2007

Restricted access

Shelby Waldron, J.D. DeFreese, Brian Pietrosimone, Johna Register-Mihalik and Nikki Barczak

were missing for 12 variables, however, missing cases did not exceed 5% for any one variable so mean imputation was utilized. Data were collected from 249 participants, however, six cases were removed from analysis due to conflicting responses on specialization grouping variables ( n  = 5) and

Restricted access

Victoria McGee and J.D. DeFreese

as a cut-off criterion for acceptability. Descriptive statistics and scale reliabilities were calculated for all study variables at every time point. Missing data from a completed assessment wave were replaced via mean imputation. Multilevel linear modeling (MLM; Singer & Willett, 2003 ) using