Examining Differential Item Functioning from a Multidimensional IRT Perspective.

Differential item functioning (DIF) is a standard analysis for every testing company. Research has demonstrated that DIF can result when test items measure different ability composites, and the groups being examined for DIF exhibit distinct underlying ability distributions on those composite abilities. In this article, we examine DIF from a two-dimensional multidimensional item response theory (MIRT) perspective. We begin by delving into the compensatory MIRT model, illustrating and how items and the composites they measure can be graphically represented. Additionally, we discuss how estimated item parameters can vary based on the underlying latent ability distributions of the examinees. Analytical research highlighting the consequences of ignoring dimensionally and applying unidimensional IRT models, where the two-dimensional latent space is mapped onto a unidimensional, is reviewed. Next, we investigate three different approaches to understanding DIF from a MIRT standpoint: 1. Analytically Uniform and Nonuniform DIF: When two groups of interest have different two-dimensional ability distributions, a unidimensional model is estimated. 2. Accounting for complete latent ability space: We emphasize the importance of considering the entire latent ability space when using DIF conditional approaches, which leads to the mitigation of DIF effects. 3. Scenario-Based DIF: Even when underlying two-dimensional distributions are identical for two groups, differing problem-solving approaches can still lead to DIF. Modern software programs facilitate routine DIF procedures for comparing response data from two identified groups of interest. The real challenge is to identify why DIF could occur with flagged items. Thus, as a closing challenge, we present four items (Appendix A) from a standardized test and invite readers to identify which group was favored by a DIF analysis.

Evaluation of the Linear Composite Conjecture for Unidimensional IRT Scale for Multidimensional Responses.

The linear composite direction represents, theoretically, where the unidimensional scale would lie within a multidimensional latent space. Using compensatory multidimensional IRT, the linear composite can be derived from the structure of the items and the latent distribution. The purpose of this study was to evaluate the validity of the linear composite conjecture and examine how well a fitted unidimensional IRT model approximates the linear composite direction in a multidimensional latent space. Simulation experiment results overall show that the fitted unidimensional IRT model sufficiently approximates linear composite direction when correlation between bivariate latent variables is positive. When the correlation between bivariate latent variables is negative, instability occurs when the fitted unidimensional IRT model is used to approximate linear composite direction. A real data experiment was also conducted using 20 items from a multiple-choice mathematics test from American College Testing.

The Effects of Gluteal Strength and Activation on the Relationship Between Femoral Alignment and Functional Valgus Collapse During a Single-Leg Landing.

CONTEXT: A bias toward femoral internal rotation is a potential precursor to functional valgus collapse. The gluteal muscles may play a critical role in mitigating these effects. OBJECTIVE: Determine the extent to which gluteal strength and activation mediate associations between femoral alignment measures and functional valgus collapse. DESIGN: Cross-sectional. SETTING: Research laboratory. PATIENTS OR OTHER PARTICIPANTS: Forty-five females (age = 20.1 [1.7] y; height = 165.2 [7.6] cm; weight = 68.6 [13.1] kg) and 45 males (age = 20.8 [2.0] y; height = 177.5 [8.7] cm; weight = 82.7 [16.5] kg), healthy for 6 months prior. INTERVENTION(S): Femoral alignment was measured prone. Hip-extension and abduction strength were obtained using a handheld dynamometer. Three-dimensional biomechanics and surface electromyography were obtained during single-leg forward landings. MAIN OUTCOME MEASURES: Forward stepwise multiple linear regressions determined the influence of femoral alignment on functional valgus collapse and the mediating effects of gluteus maximus and medius strength and activation. RESULTS: In females, less hip abduction strength predicted greater peak hip adduction angle (R2 change = .10; P = .02), and greater hip-extensor activation predicted greater peak knee internal rotation angle (R2 change = .14; P = .01). In males, lesser hip abduction strength predicted smaller peak knee abduction moment (R2 change = .11; P = .03), and the combination of lesser hip abduction peak torque and lesser gluteus medius activation predicted greater hip internal rotation angle (R2 change = .15; P = .04). No meaningful mediation effects were observed (υadj < .01). CONCLUSIONS: In females, after accounting for femoral alignment, less gluteal strength and higher muscle activation were marginally associated with valgus movement. In males, less gluteal strength was associated with a more varus posture. Gluteal strength did not mediate femoral alignment. Future research should determine the capability of females to use their strength efficiently.

Robustness of Projective IRT to Misspecification of the Underlying Multidimensional Model.

As a method to derive a "purified" measure along a dimension of interest from response data that are potentially multidimensional in nature, the projective item response theory (PIRT) approach requires first fitting a multidimensional item response theory (MIRT) model to the data before projecting onto a dimension of interest. This study aims to explore how accurate the PIRT results are when the estimated MIRT model is misspecified. Specifically, we focus on using a (potentially misspecified) two-dimensional (2D)-MIRT for projection because of its advantages, including interpretability, identifiability, and computational stability, over higher dimensional models. Two large simulation studies (I and II) were conducted. Both studies examined whether the fitting of a 2D-MIRT is sufficient to recover the PIRT parameters when multiple nuisance dimensions exist in the test items, which were generated, respectively, under compensatory MIRT and bifactor models. Various factors were manipulated, including sample size, test length, latent factor correlation, and number of nuisance dimensions. The results from simulation studies I and II showed that the PIRT was overall robust to a misspecified 2D-MIRT. Smaller third and fourth simulation studies were done to evaluate recovery of the PIRT model parameters when the correctly specified higher dimensional MIRT or bifactor model was fitted with the response data. In addition, a real data set was used to illustrate the robustness of PIRT.

Temporal kinematic differences throughout single and double-leg forward landings.

Screening methods for anterior cruciate ligament (ACL) injuries often involve double-leg landings, though the majority of ACL injuries occur during single-leg landings. Differences in kinematic temporal characteristics between single-leg and double-leg landings are poorly understood. The purpose of this study was to examine discrete and temporal kinematics associated with functional valgus collapse during single-leg and double-leg landings (LANDSL and LANDDL). Three-dimensional kinematics were obtained during the landing phases of LANDSL and LANDDL in ninety participants (45 females: 20.1 ± 1.7 yr, 165.2 ± 7.6 cm, 68.6 ± 13.1 kg; 45 males: 20.7 ± 2.0 yr, 177.7 ± 8.5 cm, 82.8 ± 16.3 kg). Peak joint angles and time series curves for frontal and transverse plane hip and knee kinematics were analyzed with an RMANOVA (discrete variables) and Statistical Parametric Mapping (SPM) paired t-tests (time series). LANDSL elicited greater knee abduction than LANDDL from 0 to 35% (0-73 ms) but greater knee adduction from 54 to 100% (112-207 ms). Peak knee abduction was 2.0° greater during LANDDL than during LANDSL (p < .001). LANDSL elicited greater hip adduction than LANDDL from 2 to 33% (4-69 ms) and greater hip abduction from 49 to 100% (102-207 ms). Peak hip adduction was 4.6° greater during LANDSL than during LANDDL (p < .001). LANDSL elicited less knee internal rotation from 0 to 3% and greater hip internal rotation from 52 to 75% of the landing phase. Peak transverse plane joint angles did not differ between tasks. During the time frame in which ACL injuries are thought to occur, LANDSL elicited frontal plane knee and hip movement consistent with risky biomechanics. Researchers and clinicians should be cognizant of how a chosen screening task alters observed kinematic effects.

Sport-specific biomechanical responses to an ACL injury prevention programme: A randomised controlled trial.

Anterior cruciate ligament (ACL) injury prevention programmes have not been as successful at reducing injury rates in women's basketball as in soccer. This randomised controlled trial (ClinicalTrials.gov #NCT02530333) compared biomechanical adaptations in basketball and soccer players during jump-landing activities after an ACL injury prevention programme. Eighty-seven athletes were cluster randomised into intervention (6-week programme) and control groups. Three-dimensional biomechanical analyses of drop vertical jump (DVJ), double- (SAG-DL) and single-leg (SAG-SL) sagittal, and double- (FRONT-DL) and single-leg (FRONT-SL) frontal plane jump landing tasks were tested before and after the intervention. Peak angles, excursions, and joint moments were analysed using two-way MANCOVAs of post-test scores while controlling for pre-test scores. During SAG-SL the basketball intervention group exhibited increased peak knee abduction angles (p = .004) and excursions (p = .003) compared to the basketball control group (p = .01) and soccer intervention group (p = .01). During FRONT-SL, the basketball intervention group exhibited greater knee flexion excursion after training than the control group (p = .01), but not the soccer intervention group (p = .11). Although women's soccer players exhibit greater improvements in knee abduction kinematics than basketball players, these athletes largely exhibit similar biomechanical adaptations to ACL injury prevention programmes.

A 6-week warm-up injury prevention programme results in minimal biomechanical changes during jump landings: a randomized controlled trial.

PURPOSE: To examine the extent to which an ACL injury prevention programme modifies lower extremity biomechanics during single- and double-leg landing tasks in both the sagittal and frontal plane. It was hypothesized that the training programme would elicit improvements in lower extremity biomechanics, but that these improvements would be greater during a double-leg sagittal plane landing task than tasks performed on a single leg or in the frontal plane. METHODS: Ninety-seven competitive multi-directional sport athletes that competed at the middle- or high-school level were cluster randomized into intervention (n = 48, age = 15.4 ± 1.0 years, height = 1.7 ± 0.07 m, mass = 59.9 ± 11.0 kg) and control (n = 49, age = 15.7 ± 1.6 years, height = 1.7 ± 0.06 m, mass = 60.4 ± 7.7 kg) groups. The intervention group participated in an established 6-week warm-up-based ACL injury prevention programme. Three-dimensional biomechanical analyses of a double- (SAG-DL) and single-leg (SAG-SL) sagittal, and double- (FRONT-DL) and single-leg (FRONT-SL) frontal plane jump landing tasks were tested before and after the intervention. Peak angles, excursions, and external joint moments were analysed for group differences using 2 (group) × 4 (task) repeated measures MANOVA models of delta scores (post-pre-test value) (α < 0.05). RESULTS: Relative to the control group, no significant biomechanical changes were identified in the intervention group for any of the tasks (n.s.). However, a group by task interaction was identified for knee abduction (λ = 0.80, p = 0.02), such that participants in the intervention group showed relative decreases in knee abduction moments during the SAG-DL compared to the SAG-SL (p = 0.005; d = 0.45, CI = 0.04-0.85) task. CONCLUSION: A 6-week warm-up-based ACL injury prevention programme resulted in no significant biomechanical changes during a variety of multi-directional jump landings. Clinically, future prevention programmes should provide a greater training stimulus (intensity, volume), more specificity to tasks associated with the mechanism of ACL injury (single-leg, non-sagittal plane jump landings), and longer programme duration (> 6 weeks) to elicit meaningful biomechanical changes. LEVEL OF EVIDENCE: I.

Biomechanical Differences of Multidirectional Jump Landings Among Female Basketball and Soccer Players.

Taylor, JB, Ford, KR, Schmitz, RJ, Ross, SE, Ackerman, TA, and Shultz, SJ. Biomechanical differences of multidirectional jump landings among female basketball and soccer players. J Strength Cond Res 31(11): 3034-3045, 2017-Anterior cruciate ligament (ACL) injury prevention programs are less successful in basketball than soccer and may be due to distinct movement strategies that these athletes develop from sport-specific training. The purpose of this study was to identify biomechanical differences between female basketball and soccer players during multidirectional jump landings. Lower extremity biomechanics of 89 female athletes who played competitive basketball (n = 40) or soccer (n = 49) at the middle- or high-school level were analyzed with 3-dimensional motion analysis during a drop vertical jump, double- (SAG-DL) and single-leg forward jump (SAG-SL), and double- (FRONT-DL) and single-leg (FRONT-SL) lateral jump. Basketball players landed with either less hip or knee, or both hip and knee excursion during all tasks (p ≤ 0.05) except for the SAGSL task, basketball players landed with greater peak hip flexion angles (p = 0.04). The FRONT-SL task elicited the most distinct sport-specific differences, including decreased hip adduction (p < 0.001) angles, increased hip internal rotation (p = 0.003), and increased relative knee external rotation (p = 0.001) excursions in basketball players. In addition, the FRONT-SL task elicited greater forces in knee abduction (p = 0.003) and lesser forces in hip adduction (p = 0.001) and knee external rotation (p < 0.001) in basketball players. Joint energetics were different during the FRONT-DL task, as basketball players exhibited less sagittal plane energy absorption at the hip (p < 0.001) and greater hip (p < 0.001) and knee (p = 0.001) joint stiffness. Sport-specific movement strategies were identified during all jump landing tasks, such that soccer players exhibited a more protective landing strategy than basketball players, justifying future efforts toward sport-specific ACL injury prevention programs.

Effects of exercise accumulation on plasma lipids and lipoproteins.

Debate exists as to whether improvements in some cardiometabolic risk factors following exercise training result more from the last session of, or from an accumulation of, exercise sessions. This study was designed to compare the effect of a single exercise session with 3 consecutive days of exercise on triglyceride (TG), high-density lipoprotein-cholesterol (HDL-C), and low-density lipoprotein-cholesterol (LDL-C). Twelve young adult (aged 22.5±2.5 years), overweight (body mass index=29.7±4 kg·m(-2)), sedentary, black (n=5) and white (n=7) men (n=6) and women (n=6) completed, in random order, a single treadmill exercise session at 60% maximal oxygen uptake for 90 min (1EX), accumulated exercise sessions (same as for 1EX) for 3 consecutive days (3EX), and a control protocol (no exercise for 6 days). Plasma samples were collected from baseline through 24, 48, and 72 h postexercise. Significant treatment-by-time interactions (p<0.05) existed in HDL-C and LDL-C. Postexercise responses of HDL-C differed at 48 h (1EX: -3.6, 3EX: +3.7 mg·dL(-1)) and 72 h (1EX: -1.7, 3EX: +3.2 mg·dL(-1)). LDL-C responses differed at 48 h (1EX: -16, 3EX: +6 mg·dL(-1)). Although not statistically significant, TG concentrations decreased by 29% at 24 h after 3EX, compared with -7% after 1EX. An inverse relationship between baseline and postexercise reduction in TG was present with 3EX (r=-0.655; p<0.05). In conclusion, 3EX increased HDL-C and decreased TG more than 1EX, while the decrease in LDL-C after 1EX was suppressed. Blood lipid panel changes may be due to more accumulated effects over time rather than just a result of the most recent exercise session.

Lower body stiffness and muscle activity differences between female dancers and basketball players during drop jumps.

BACKGROUND: Anterior cruciate ligament (ACL) injuries often occur during landing, with female athletes at higher injury risk than male athletes. Interestingly, female dancers have lower ACL injury rates than do female athletes in general. HYPOTHESIS: Female dancers will have earlier and greater lower extremity muscle activity and higher sagittal knee joint and leg stiffness than will female basketball players. STUDY DESIGN: Cross-sectional group comparison. METHODS: Fifty-five healthy female athletes (35 dancers, 20 basketball players) performed 5 double-leg drop jumps from a 45-cm box. Surface electromyography (onsets and amplitudes; prelanding and postlanding) was recorded from the lateral gastrocnemius, medial and lateral hamstrings, lateral quadriceps muscles with a 3-dimensional electromagnetic tracking system, and forceplates recording biomechanics (leg spring stiffness and knee joint stiffness). RESULTS: Compared with basketball players, dancers had greater leg spring stiffness (P = 0.047) but similar knee joint stiffness (P = 0.44). Although no significant differences were observed in overall muscle onset times (P = 0.22) or activation amplitudes (prelanding, P = 0.60; postlanding, P = 0.78), small to moderate effect sizes (ESs) suggest trends in dancers toward earlier (ES = 0.53) and higher medial hamstrings activation pre- (ES = 0.55) and post- (ES = 0.41) landing and lower lateral quadriceps (ES = 0.30) and higher gastrocnemius (ES = 0.33) postlanding muscle activation. CONCLUSIONS: In dancers, the higher leg spring stiffness and trends toward higher hamstrings prelanding and postlanding, as well as lower quadriceps and higher gastrocnemius activation postlanding with similar knee joint stiffness, indicate lower extremity neuromechanical differences across other joints. CLINICAL RELEVANCE: Female dancers may have lower extremity neuromechanics that are different from those of basketball players during drop jumps. If dancers use ACL-protective strategies during activity, then their training routines should be further investigated to improve ACL injury prevention programs.