Search Results

You are looking at 11 - 20 of 74 items for :

  • "imputation" x
  • Psychology and Behavior in Sport/Exercise x
Clear All
Restricted access

Oleg Zaslavsky, Yan Su, Eileen Rillamas-Sun, Inthira Roopsawang and Andrea Z. LaCroix

strength was measured by a handgrip dynamometer and was rounded up to the nearest kilogram (in kg). BMI was calculated by dividing the weight (in kilograms) by the square of the height (in meters). Missing Data We conducted multiple imputations based on maximizing expectation method to account for the

Restricted access

Tiago V. Barreira, Stephanie T. Broyles, Catrine Tudor-Locke, Jean-Philippe Chaput, Mikael Fogelholm, Gang Hu, Rebecca Kuriyan, Estelle V. Lambert, Carol A. Maher, José A. Maia, Timothy Olds, Vincent Onywera, Olga L. Sarmiento, Martyn Standage, Mark S. Tremblay, Peter T. Katzmarzyk and for the ISCOLE Research Group

chance of bias due to exclusion of these cases. Missing values were multiply imputed (5 imputations) using fully conditional specification methods, under missing at random assumptions 24 and using SAS software (version 9.4; PROC MI, SAS Institute Inc., Cary, NC). Country-specific models were used to

Restricted access

Aubrey Newland, Rich Gitelson and W. Eric Legg

indicated that data were missing completely at random, χ 2  = 359.151, df  = 321, p  = .07, and <5% of the data were missing. Thus, hot-deck imputation ( Myers, 2011 ) was used to create the final data set. Hot-deck imputation is a method of replacing missing values of the nonrespondent with data from a

Restricted access

Shijun Zhu, Eun-Shim Nahm, Barbara Resnick, Erika Friedmann, Clayton Brown, Jumin Park, Jooyoung Cheon and DoHwan Park

). Participants who dropped out had frailer health (i.e., more arthritis, depression) than those who stayed in the study and were more likely to be hospitalized during the past 3 months prior to the baseline survey. We chose FIML instead of multiple imputation as FIML showed more efficiency in estimation than

Restricted access

David Geard, Amanda L. Rebar, Peter Reaburn and Rylee A. Dionigi

rate their life these days (0 =  the worst possible life to 10 =  the best possible life ) ( Pruchno & Wilson-Genderson, 2014 ). Data Analyses Model fit analyses Multiple imputation by chained equations was used to account for missing data in this study because it imputes complex multivariate data by

Restricted access

David X. Marquez, Robert Wilson, Susan Aguiñaga, Priscilla Vásquez, Louis Fogg, Zhi Yang, JoEllen Wilbur, Susan Hughes and Charles Spanbauer

than multiple imputation. Because it resamples the empirical dataset, rather than fitting it to a normal distribution, it better retains the shape of non-normal distributions. One disadvantage of hot decking is that while it retains the same mean and variance as the original data, it can change the

Restricted access

Janet M. Boekhout, Brenda A.J. Berendsen, Denise A. Peels, Catherine A.W. Bolman and Lilian Lechner

point, and an interaction term between the weekly minutes of MVPA and measurement time point. Imputation of missing data was not applied, as it has been established that the use of incomplete cases in multilevel analysis results in more accurate estimations than applying imputation ( Twisk, De Boer, De

Restricted access

Keith R. Lohse, Jincheng Shen and Allan J. Kozlowski

complicated statistical adjustments (such as multiple imputation) when data are missing at random. • Including only cases with complete data will reduce statistical power and risk bias to the model if data are missing not at random (MNAR). • Depending on the method employed, imputing missing values may not

Restricted access

Natalie Kružliaková, Paul A. Estabrooks, Wen You, Valisa Hedrick, Kathleen Porter, Michaela Kiernan and Jamie Zoellner

sex, age, race, ethnicity, educational attainment, employment/disability status, income, health literacy level, number of children, smoking status, and BMI. In addition, models used intention-to-treat with baseline observation carried forward imputations. 36 , 54 , 55 Results Demographics Of the 301

Restricted access

Kari Roethlisberger, Vista Beasley, Jeffrey Martin, Brigid Byrd, Krista Munroe-Chandler and Irene Muir

screened to assess missing data and normality. Missing data is typically not a problem if it is under 5% or 10% ( Bennett, 2001 ; Schafer, 1999 ) Mean imputation was used for seven missing data points for five variables. Given the low percentage (.002%) of missing data this approach is considered