Measuring Microaggressions in Youth from Three Major Racial/Ethnic Groups

Background/Purpose:

According to the National Assessment of Educational Progress, only one third of American 8th graders demonstrated proficiency in mathematics and reading in 2015. Equally alarming are differences in proficiency rates across racial/ethnic groups. Mathematics and reading proficiency rates for Black and Latino 8^th graders ranged from 13% to 21%, compared to over 40% for Whites. The statistics suggests an urgent need for efforts to identify and address potential barriers to learning encountered in schools by students of color.

Racial/ethnic “microaggressions” (REMs) are theoretically defined as subtle, recurrent, and repetitive forms of racism and discrimination experienced by people of color. REMs may be conscious or unconscious, explicit or implicit; however, they affect the mental health and daily functioning of targeted individuals. Most research on REMs has focused on adults, leaving a gap in knowledge about REM experiences of adolescents and their effects on well-being and outcomes. A first step in understanding REMs among youth is to develop or identify a high-quality, adolescent measure. This study examines the performance as a measure of REMs of an existing scale on a well-validated tool for adolescents.

Methods:

Data/Sample: Data used in this cross-sectional study were collected from 2009 to 2014 from 9536 middle and high school students in 66 schools in 8 states. Data were collected with the School Success Profile (SSP), an online, self-report, social environmental assessment developed specifically for adolescents. Our study focused on 8215 cases meeting our age (<18) and race/ethnicity (Black, Latino, White) criteria.

Measures: The SSP Microinteraction scale comprises 13 items assessing students’ perceptions of negative interactions at school. The items’ ordinal response options are: “never,” “once or twice,” and “more than twice.”

Analysis: We conducted a recommended sequence of invariance tests to assess the quality of the Microinteraction scale and its performance across three major racial and ethnic groups. We used MPlus 7.4’s invariance “shortcut,” and the appropriate estimator (WLSMV) and analysis matrix (polychoric correlations) for ordinal data. The model was developed and confirmed on two randomly generated samples. Rigorous fit criteria were applied. Construct validity tests were also performed.

Results: Items have face validity because they reflect a common taxonomy of REMs. Individual-group CFAs indicated the same one-factor, 13-item solution for all groups. Invariance tests revealed the measure had good fit and was invariant for Black and Latino youth. Twelve of 13 factor loadings were non-invariant for White youth, however, suggesting the construct of microaggressions for Whites is defined differently. Furthermore, good fit of the final model for Whites was not replicated in the confirmation sample. As hypothesized, scale scores were associated with youth behavior, engagement, and mental health outcomes.

Conclusions and Implications

Results suggest the SSP Microinteraction scale may be used as a measure of REMs among Black and Latino middle and high school students. The appropriateness of the measure for White youth is questionable, which increases confidence that it taps racial/ethnic experiences. Advantages of the existing SSP scale include its brevity, school focus, and its association with a well-validated and commonly used