Evaluating Measurement Invariance with Censored Ordinal Data: A Monte Carlo Comparison of Alternative Model Estimators and Scales of Measurement

Evaluations of measurement invariance provide essential construct validity evidence. However, the quality of such evidence is partly dependent upon the validity of the resulting statistical conclusions. The presence of Type I or Type II errors can render measurement invariance conclusions meaningless. The purpose of this study was to determine the effects of categorization and censoring on the behavior of the chi-square/likelihood ratio test statistic and two alternative fit indices (CFI and RMSEA) under the context of evaluating measurement invariance. Monte Carlo simulation was used to examine Type I error and power rates for the (a) overall test statistic/fit indices, and (b) change in test statistic/fit indices. Data were generated according to a multiple-group single-factor CFA model across 40 conditions that varied by sample size, strength of item factor loadings, and categorization thresholds. Seven different combinations of model estimators (ML, Yuan-Bentler scaled ML, and WLSMV) and specified measurement scales (continuous, censored, and categorical) were used to analyze each of the simulation conditions. As hypothesized, non-normality increased Type I error rates for the continuous scale of measurement and did not affect error rates for the categorical scale of measurement. Maximum likelihood estimation combined with a categorical scale of measurement resulted in more correct statistical conclusions than the other analysis combinations. For the continuous and censored scales of measurement, the Yuan-Bentler scaled ML resulted in more correct conclusions than normal-theory ML. The censored measurement scale did not offer any advantages over the continuous measurement scale. Comparing across fit statistics and indices, the chi-square-based test statistics were preferred over the alternative fit indices, and ΔRMSEA was preferred over ΔCFI. Results from this study should be used to inform the modeling decisions of applied researchers. However, no single analysis combination can be recommended for all situations. Therefore, it is essential that researchers consider the context and purpose of their analyses.

Evaluation of an In Vitro Blood-Based Assay to Detect Production of Interferon-γ By Mycobacterium Bovis–Infected White-Tailed Deer (Odocoileus Virginianus)

Tuberculosis due to Mycobacterium bovis in captive Cervidae was identified as an important disease in the United States in 1990 and prompted the addition of captive Cervidae to the USDA Uniform Methods and Rules for eradication of bovine tuberculosis. As well, M. bovis infection was identified in free-ranging white-tailed deer in northeast Michigan in 1995. Tuberculosis in both captive and free-ranging Cervidae represents a serious challenge to the eradication of M. bovis infection from the United States. Currently, the only approved antemortem tests for tuberculosis in Cervidae are the intradermal tuberculin skin test and the blood tuberculosis test (BTB). At present, the BTB is not available in North America. Tuberculin skin testing of Cervidae is time-consuming and involves repeated animal handling and risk of injury to animals and humans. This study evaluated the potential of a new blood-based assay for tuberculosis in Cervidae that would decrease animal handling, stress, and losses due to injury. In addition, a blood-based assay could provide a more rapid diagnosis. Twenty 6–9-month-old white-tailed deer, male and female, were experimentally inoculated by instillation of 300 colony-forming units of M. bovis in the tonsillar crypts. Seven, age-matched uninfected deer served as controls. Blood was collected on days 90, 126, 158, 180, 210, 238, 263, and 307 after inoculation and was analyzed for the production of interferon-γ (IFN-γ) in response to incubation with M. bovis purified protein derivative (PPDb), M. avium PPDa, pokeweed mitogen (PWM), or media alone. Production of IFN-g in response to PPDb was significantly greater (P < 0.05) at all time points in samples from M. bovis–infected deer as compared with uninfected control deer, whereas IFN-γ production to PWM did not differ significantly between infected and control deer. Measurement of IFN-γ production to PPDb may serve as a useful assay for the antemortem diagnosis of tuberculosis in Cervidae.

Shared Feed as a Means of Deer-to-Deer Transmission of Mycobacterium bovis

To determine the ability of experimentally inoculated white-tailed deer (Odocoileus virginianus) to transmit Mycobacterium bovis to naive deer through the sharing of feed, four deer were intratonsillarly inoculated with 4x105 colony-forming units of M. bovis. On a daily basis, feed not consumed by inoculated deer after approximately 8 hr was offered to four naıve deer maintained in a separate pen, where direct contact, aerosol transmission, or transmission through personnel were prevented. After 150 days, naıve deer were euthanized and examined. All naıve deer had lesions consistent with tuberculosis and M. bovis was isolated from various tissues. The most commonly affected tissues were lung, tracheobronchial lymph nodes, and mediastinal lymph nodes. This study demonstrates the potential for indirect transmission of M. bovis through the sharing of feed. Intentional or unintentional feeding of deer by wildlife or agricultural interests in regions where M. bovis infection is endemic should be avoided because both direct and indirect transmission through sharing of feed are enhanced.