220 The Professional Counselor | Volume 12, Issue 3 their privacy and confidentiality; however, authors can give descriptions of the setting and location (e.g., “Data were collected between April 2021 to February 2022 from clients seeking treatment for addictive disorders at an outpatient, integrated behavioral health care clinic located in the Northeastern United States.”). Authors should also report details about any interventions, curriculum, qualifications and background information for research assistants, experimental design protocol(s), and any other procedural design issues that would be necessary for replication. In instances in which describing a treatment or conditions becomes exorbitant (e.g., step-by-step manualized therapy, programs, or interventions), researchers can include footnotes, appendices, and/or references to refer the reader to more information about the intervention protocol. Missing Data Procedures for handling missing values (incomplete survey responses) are important considerations in quantitative data analysis. Perhaps the most straightforward option for handling missing data is to simply delete missing responses. However, depending on the percentage of data that are missing and how the data are missing (e.g., missing completely at random, missing at random, or not missing at random), data imputation techniques can be employed to recover missing values (Cook, 2021; Myers, 2011). Quantitative researchers should provide a clear rationale behind their decisions around the deletion of missing values or when using a data imputation method. Readers are encouraged to review Cook’s (2021) commentary on procedures for handling missing data in quantitative research. Measures Counseling and other social science researchers oftentimes use instruments and screening tools to appraise latent traits, which can be defined as variables that are inferred rather than observed (AERA et al., 2014). The purpose of the Measures (aka Instrumentation) section is to operationalize the construct(s) of measurement (Heppner et al., 2016). Specifically, the Measures subsection of the Methods in a quantitative manuscript tends to include a presentation of (a) the instrument and construct(s) of measurement, (b) reliability and validity evidence of test scores, and (c) cross-cultural fairness and norming. The Measures section might also include a Materials subsection for studies that employed data-gathering techniques or equipment besides or in addition to instruments (Heppner et al., 2016); for instance, if a research study involved the use of a biofeedback device to collect data on changes in participants’ body functions. Instrument and Construct of Measurement Begin the Measures section by introducing the questionnaire or screening tool, its construct(s) of measurement, number of test items, example test items, and scale points. If applicable, the Measures section can also include information on scoring procedures and cutoff criterion; for example, total score benchmarks for low, medium, and high levels of the trait. Authors might also include commentary about how test scores will be operationalized to constitute the variables in the upcoming Data Analysis section. Reliability and Validity Evidence of Test Scores Reliability evidence involves the degree to which test scores are stable or consistent and validity evidence refers to the extent to which scores on a test succeed in measuring what the test was designed to measure (AERA et al., 2014; Bardhoshi & Erford, 2017). Researchers should report both reliability and validity evidence of scores for each instrument they use (Wester et al., 2013). A number of forms of reliability evidence exist (e.g., internal consistency, test-retest, interrater, and alternate/parallel/ equivalent forms) and the AERA standards (2014) outline five forms of validity evidence. For the purposes of this article, I will focus on internal consistency reliability, as it is the most popular and most
RkJQdWJsaXNoZXIy NDU5MTM1