Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Jan 1.
Published in final edited form as: Meas Eval Couns Dev. 2023 Apr 25;57(1):1–14. doi: 10.1080/07481756.2023.2193339

Development and Initial Validation of Scores on The Lifestyle Practices and Health Consciousness Inventory-2: Brief Version

Michael T Kalkbrenner 1, Michaela Bradley 1, Hannah Sun 1
PMCID: PMC10756427  NIHMSID: NIHMS1932376  PMID: 38164162

Abstract

We developed and validated scores on the Lifestyle Practices and Health Consciousness Inventory (LPHCI)-2: Brief Version, a short form for measuring global wellness (mental and physical health). Tests of internal structure (EFA, CFA, and higher-order CFA) as well as convergent validity supported the psychometric properties of LPHCI-2: Brief Version scores.

Keywords: Integrated primary care, lifestyle practices, wellness, health consciousness

Development, prevention, and wellness are deeply rooted in the professional identity of counselors (Myers, 1992). Accordingly, counselors are uniquely positioned to address the growing prevalence and complexity of mental health issues and non-communicable physical disease (Center for Disease Control and Prevention [CDC], 2022; National Alliance on Mental Illness [NAMI], 2022). The NAMI (2022) estimates that approximately one in every five adults in the United States struggles with a mental health issue(s) each year. In addition, non-communicable diseases, for example cardiovascular diseases, are a leading cause of death among adults living in the United States (CDC, 2022). The early detection of symptoms of mental and physical health distress via screening tools plays a crucial role in connecting clients with support services before their symptoms develop into serious health issues. Consistent with the wellness orientation of the counseling field (Myers, 1992), integrated primary care (IPC) is a nascent healthcare trend in the United States, which involves strategic collaboration between medical and mental healthcare providers to provide holistic patient care (McCauley et al., 2021; Mulvaney-Day et al., 2018). The emerging popularity of IPC is due to its many advantages, including but not limited to its utility for simultaneously promoting patients’ physical and mental wellness, early identification and prevention of serious health care issues, and potential to save the healthcare system billions of dollars each year (Jimenez-Lara, 2016; McCauley et al., 2021).

Professional counselors and other IPC practitioners need screening tools to establish baseline health levels, aid in treatment planning, and to monitor their clients’ progress toward meeting the goals of their treatment plans (Mulvaney-Day et al., 2018). Particularly, brief versions of screening tools (10 or fewer items) have utility in IPC settings (Mulvaney-Day et al., 2018; Shields et al., 2021). The Lifestyle Practices and Health Consciousness Inventory (LPHCI) is a screening tool for simultaneously appraising mental and physical health with a number of rigorous score validation studies to support its psychometric properties (Kalkbrenner & Gormley, 2020; Kalkbrenner, 2022a; 2022b). However, the LPHCI is too lengthy to incorporate into the already extensive intake screening process in most medical and mental health settings (Kleiman et al., 2020; McDermott et al., 2019). Accordingly, the aim of the present study was to develop and validate scores on the LPHCI-2: Brief Version with a national stratified random sample of adults living in the United States. If scores are validated, professional counselors and other members of IPC teams can use scores on the LPHCI-2: Brief Version to screen clients in the context of a primary care visit to help determine if a referral to another provider might be helpful.

Total Wellness Screening Tools in Medical and Mental Health Settings

The American Counseling Association (ACA, 2014) encourages counselors to use screening tools to aid in the detection, diagnosis, and treatment of their clients’ presenting concerns. The Five Factor Wellness Inventory (5F-Wel) is a comprehensive screening tool for appraising total wellness, “a measure of one’s general well-being” (Myers & Sweeney, 2014, p. 10). A substantial body of empirical support exists for the psychometric properties of the 5F-Wel (Shannonhouse et al., 2020). However, two of the primary barriers to using the 5F-Wel in IPC settings include, (a) the length of the instrument (70+ items) and (b) its use is proprietary, which some agencies or independent practitioners cannot afford on a regular basis.

In an extension of the wellness-based measurement literature, Kalkbrenner and Gormley (2020) developed and validated scores on the LPHCI for appraising global wellness (overall mental and physical health). The LPHCI begins to address the logistical barriers of the 5F-Wel, as the LPHCI it is free to use and comprised of far fewer items (i.e., 20). In addition, the LPCHI consists of four dimensions or subscales that appraise four separate dimensions of total wellness. The four subscales can be used separately, however, one of the most notable contributions of the LPHCI to the measurement literature is the Global Wellness scale, a higher-order factor that produces a composite total wellness score, which is comprised of all four subscales. However, the already lengthy intake test batteries in most IPC settings makes adding an additional 20-item survey unfeasible. In fact, a 20-item screening tool is more than twice as long as the recommended length for short forms in psychological research (Shields et al., 2021). Thus, there is potential to develop a brief version of the LPHCI with the goal of capturing global wellness with 10 or fewer items.

Utility of Brief Screening Tools in Medical and Mental Health Settings

Brief screening tools (aka short forms) with approximately 10 or fewer items have utility in IPC considering the lengthy intake assessment protocols that exist in these settings (McCauley et al., 2021; Schipolowski et al., 2014; Shields et al., 2021). In fact, some of the most widely used screening tools in healthcare settings are short forms comprised of < 10 items, for example, The Patient Health Questionnaire- 9 (Kroenke et al., 2001), Generalized Anxiety Disorder-7 (Spitzer et al., 2006), and Mental Health Inventory-5 (Berwick et al., 1991). To the best of our knowledge, the literature is missing a brief total wellness screening tool with validated scores, which is problematic considering the growing IPC healthcare trend in the United States. Accordingly, the development and validation of scores on a brief version of the LPHCI has potential to make a major contribution to the measurement literature. If scores are validated, professional counselors and other IPC practitioners can use the LPHCI-2: Brief Version to screen clients in the context of a primary care visit, which might help them get connected with additional treatment. In this spirit, the following research questions (RQs) guided the present investigation: RQ1: What is the dimensionality of scores on a newly developed brief version of the LPHCI with a national sample of adults in the United States? RQ2: Is the emergent LPHCI-2: Brief Version factor structure confirmed with a second sample of adults in the United States? RQ3: What is the convergence of scores between the LPHCI-2: Brief Version with the full version of the LPHCI?

Methods

Instrument Development Process

The LPHCI-2: Brief Version was developed based on the MEASURE Approach to instrument development and score validation (Kalkbrenner, 2021). MEASURE is an acronym comprised of the following empirically supported steps for conducting an instrument development and score validation study: Make the purpose and rationale clear, establish empirical framework, articulate theoretical blueprint, synthesize content and scale development, use expert reviewers, recruit participants, and evaluate validity and reliability evidence of scores (Kalkbrenner, 2021). Based on the MEASURE Approach, we used a research team throughout the instrument development and score validation process. The research team consisted of an associate professor of counseling, a graduate student in clinical mental health counseling, and a graduate student in school psychology. The first decision the research team considered was whether the LPHCI-Brief Version would be comprised of a reduced pool of the original LPHCI items developed by Kalkbrenner and Gormley (2020) or if a new pool of items (a different version) would be created.

When developing a brief version of a screening tool, researchers are tasked with sustaining the scope and depth of the latent trait with the fewest possible number of items (Clark & Watson, 2019; Shields et al., 2021). Accordingly, item development involves navigating a delicate balance between writing items with enough specificity for clarity of meaning while also covering a wide scope of content (Clark & Watson, 2019). The research team elected to develop a second version (new pool of items) for the LPHCI-Brief, due to the specific content of the original LPHCI items. However, the dimensionality of the original LPHCI still served as the theoretical framework for the brief version.

The research team worked together to design a theoretical blueprint with content areas consisting of the four original LPHCI dimensions (Aerobic Exercise, Food Choices, Self-Care, and Consciousness of Stress). Following the MEASURE approach, each research team member referred to the blueprint and the established definition of global wellness: “(a) lifestyle practices, or activities, routines, and dietary habits, which are related to physical and/or mental wellness, and (b) health consciousness, or the extent to which people are aware of their own mental and physical health” (Kalkbrenner, 2022a, p. 85) to develop separate lists of possible LPHCI-2: Brief Version items. Subsequently, the research team came together to review and discuss each members’ list of possible items and eventually reached a consensus on the original pool of 45 items for the upcoming expert review phase. The original LPHCI prompt and response scale points (Kalkbrenner & Gormley, 2020) was used for the LPHCI-2: Brief Version: “In the past 30 days, how often have you…,” on the following scale, 0 = never, 1 = one to five times, 2 = six to ten times, 3 = eleven to fifteen times, 4 = sixteen to twenty times, or 5 = twenty-one or more times.” (p. 222).

Based on the MEASURE approach, we sent the original 45 LPHCI-2: Brief Version items to three expert reviewers, including one survey expert and two content experts (Kalkbrenner, 2021). The first content expert is the director of a health and wellness institute, which focuses on simultaneously improving clients’ physical health through exercise and proper diet as well as their mental health through therapy. The second content expert has a Ph.D. in counseling psychology, with over 10 years of experience working in an integrated behavioral health setting. The survey expert has a Ph.D. in counselor education with an extensive background (20+ years) in testing, assessment, and psychometrics. The expert reviewers suggested removing seven items due to redundancy and unclear phrasing. Among the remaining 38 items, the subject-matter experts did not suggest any additional construct attributes nor did they identify any irrelevant items.

Participants and Procedures

The 38 LPHCI-2: Brief Version items were entered into the Qualtrics online survey platform for pilot testing. The authors were awarded an Institutional Development Award (IDeA) grant from the National Institute of General Medical Science to fund data collection. Grant funding was used to hire Qualtrics Sample Services (2022; a data collection contractor). Following approval from the institutional review board, the corresponding author emailed the anonymous electronic distribution link to a project manager with Qualtrics Sample Services for pilot testing. Pilot studies tend to include between 25 and 150 participants (Browne, 1995; Hertzog, 2008). Qualtrics Sample Services (2022) made the pilot study items available to eligible participants and stopped data collection after receiving 52 completed responses. In addition to responding to the test items, pilot study participants were given the opportunity to offer feedback on the content, readability, and/or relevance of the items for appraising global wellness. Pilot study participants did not suggest any changes to the items. A review of the pilot data revealed no technology or data imputation errors.

Following the lead author’s review and approval of the pilot data, the Qualtrics project management team recruited a national random sample, stratified by the U.S. Census data (2022) for ethnicity, age, gender, and geographic location of adults living in the United States. A team of analysts from Qualtrics Sample Services conducted a quality check on the data set to identify and remove potential poor-quality responses, which included an attention check. A total of 844 quality responses were collected with 0% missing data. The data were randomly divided into two samples including an EFA sample (N = 544) and a CFA (N = 300) sample.

EFA Sample

A review of the EFA data set exposed 43 univariate outliers (z > −/+ 3.29; Field, 2018), which were removed from the data set, yielding a robust EFA sample of N = 501. A review of skewness and kurtosis values in the EFA data set revealed no extreme deviations from normality (skewness values > −/+2 and kurtosis > −/+7, Dimitrov, 2012), with no absolute skewness values exceeding 1.43, and no absolute kurtosis values exceeding 1.47. The minimum sample size for factor analysis should include at least 200 participants or a participants-to-variables ratio of 10:1, whichever produces a larger sample (Kalkbrenner, 2021). The participants-to-variables ratio (13:1) for the final EFA sample of (N = 501), exceeded both 10:1 and 200 participants.

Participants in the EFA sample ranged in age from 18 to 88 years (M = 49.19, SD = 17.89). For gender identity, 51.1% (n = 254) of participants self-identified as female, 47.9% (n = 238) male, 0.6% (n = 3) transgender, 0.4% (n = 2) non-binary, and 0.7% (n = 4) preferred not to specify their gender identity. For ethnic identity, 63.6% of the participants identified as White or European American (n = 316), 17.3% as Hispanic, Latinx, or Spanish origin (n = 86), 8% as Black or African American (n = 40), 5.4% as Asian or Asian American (n = 27), 3% as multiethnic (n = 15), 1% as Middle Eastern or North African (n = 5), 1% as American Indian or Alaska Native (n = 5), 0.4% as Native Hawaiian or Other Pacific Islander, and 0.2% identified as Another race, ethnicity, or origin (please specify; n = 1). For region of the United States, 34.8% (n = 173) were from the South, 27.4% (n = 136) from the West, 22.1% (n = 110) from the Midwest, 15.7% (n = 78) from the Northeast, and 0.7% (n = 4) preferred not to specify their geographic location.

CFA Sample

A review of the CFA sample (N = 300) revealed zero univariate outliers (z > −/+ 3.29) and Mahalanobis (D) distance revealed an absence of multivariate outliers. Skewness and kurtosis values were consistent with a normal distribution with no absolute skewness values exceeding 0.30, and no absolute kurtosis values exceeding 1.39. Participants in the CFA sample ranged in age from 18 to 83 (M = 43.69, SD = 17.23). For gender identity, 55.3% participants identified as female (n = 166), 44% as male, (n = 132), and 0.7% identified as non-binary (n = 2). For ethnic identity, 44.3% identified as White or European American (n = 133), 23% as Black or African American (n = 69), 19.3% as Hispanic, Latinx, or Spanish origin (n = 58), 5.3% as Asian or Asian American (n = 16), 2.7% as multiethnic (n = 8), 2% as American Indian or Alaskan Native (n = 6), 1.3% as another race, ethnicity, or origin (please specify; n = 4), 1% as Middle Eastern or North African (n = 3), and 1% said I prefer not to say (n = 3). For region, 42.7% (n = 128) were from the South, 19.7% (n = 59) from the Northeast, 19% (n = 57) from the West, and 18.7% (n = 56) from the Midwest.

Measures

Participants gave their informed consent, confirmed that they met the eligibility criteria (18 years or older and living in the United States), and responded to a demographic questionnaire. The demographic questionnaire included items regarding participants’ age, gender identity, ethnic identity, and geographic location. Participants then completed the LPHCI-2: Brief Version (see Appendix A). Finally, participants completed the original LPHCI for convergent validity testing purposes.

Life-Style Practices and Health Consciousness Inventory

The LPHCI is a screening tool to evaluate global wellness, which Kalkbrenner and Gormley (2020) defined as physical and mental health supported by lifestyle practices such as activities, routines, dietary habits, and having health consciousness which is an awareness of their own mental and physical health. The original version LPHCI consists of 20 items that focused on Consciousness of Stress (one’s awareness of their own mental health and physical health), Self-Care (engagement in fulfilling and relaxing activities), Aerobic Exercise (participation in physical activities), Food Choices (dietary habits and nutrition), and includes a higher-order Global Wellness scale (Kalkbrenner & Gormley, 2020).

An emergent body of literature supports reliability and validity evidence of LPHCI scores. Past investigators established internal structure validity (CFAs and multiple-group CFAs) as well as convergent validity evidence (rs > .50) of LPHCI scores (Kalkbrenner & Gormley, 2020; Kalkbrenner 2022a; 2022b). In addition, the authors of a number of previous LPHCI studies found acceptable-to-strong internal consistency reliability estimates of scores on the Global Wellness scale of the LPHCI, α = .92 and ωh = .83 (Kalkbrenner, 2022b), α = .94 (Kalkbrenner, 2022a), and α = .92 (Kalkbrenner & Gormley, 2020).

Guidelines for Interpreting Validity and Reliability Evidence of Scores

Measuring reliability and validity evidence of scores is a complex pursuit. Accordingly, it is best practice to investigate multiple fit indices and interpret the collective package of the results (Dimitrov, 2012). The combined recommendations of leading psychometric researchers (Dimitrov, 2012; Schreiber et al., 2006) for interpreting goodness of fit indices were used to evaluate internal structure validity of scores, the Chi square absolute fit index (CMIN, non-significant p-value), the comparative fit index (CFI, .90 to .95 = acceptable fit and > .95 = strong fit), the Tucker-Lewis index (TLI, .90 to .95 = acceptable fit and > .95 = strong fit, root mean square error of approximation (RMSEA < .08 and < .06 = strong fit), and standardized root mean square residual (SRMR < .08 and < .06 = strong fit). Convergent validity evidence was tested through bivariate correlations, with high correlations (r > .50; Drummond et al., 2016) denoting convergent validity evidence of scores.

Multiple indices were also interpreted to investigate the internal consistency reliability evidence of LPHCI-2: Brief Version scores. First, we computed Cronbach’s coefficient alpha (α), as it is the most widely used internal consistency reliability estimate in the social sciences literature (Kalkbrenner, 2021; McNeish, 2018). However, the stability of alpha is dependent on the data meeting multiple assumptions, which are oftentimes violated (especially in psychometric research). Accordingly, we also computed more robust estimates of internal consistency reliability, including McDonald’s coefficient omega (ω) and coefficient H. Coefficient omega is not vulnerable to most of α’s assumptions (i.e., ω does not get distorted in the same ways as α when the data violate one or more of α’s assumptions). Coefficient H is based on optimal scale weighting and yields a maximum reliability estimate. The composite score is not adversely impacted by individual items that contribute less stability in the way that α and ω are (Kalkbrenner, 2021; McNeish, 2018). We used the following guidelines for acceptable internal consistency reliability estimates of scores, α ≥ .70 (Tavakol & Dennick, 2011), ω ≥ .65 (Nájera Catalán, 2019), and H ≥ .70 (Nunnally & Bernstein, 1994).

Results

A series of psychometric tests were employed based on the MEASURE approach to establish construct validity evidence of scores on the LPHCI-2: Brief Version (Kalkbrenner, 2021). Psychometric testing began with internal structure validity testing, including both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA). Evidence of convergent validity was investigated by correlating scores on the LPHCI and LPHCI-2.

Exploratory Factor Analysis

The three following preliminary EFA analyses were computed based on the recommendations of Mvududu and Sink (2013): Bartlett’s test of sphericity, Kaiser-Meyer-Olkin measure of sampling adequacy, and inter-item correlation matrix. All items displayed correlations r < .20 with at least three other items. The results of Bartlett’s test of sphericity (B [741] = 10842.10, p < .001) and the Kaiser-Meyer-Olkin measure of sampling adequacy (KMO = .92) revealed that the data were factorable.

The 38 LPHCI-2: Brief Version items were entered into an EFA with principal axis factoring. The following combined guidelines from Mvududu and Sink (2013) and Beavers et al. (2013) for factor retention were used: factor loadings > 0.40 and communalities (h2) > .30. Cross-loadings were identified by values greater than 0.35, which were eliminated. The initial factor extraction based on the Kaiser criterion revealed an initial 7-factor solution, which accounted for 62.1% of the variance in the model. Three factor extraction criteria revealed different factor solutions, including the Scree plot (4 factors), meaningful variance accounted for (> 5%; 3 factors), and parallel analysis (6 factors). Each factor solution was tested with an oblique rotation (direct oblimin), as total wellness-based constructs tend to inter-correlate (Kalkbrenner & Gormley, 2020). Items that displayed h2 values < .30, failed to load on any factor, or cross-loaded on multiple factors were removed one at a time. The EFA was recomputed after each removal. The six factor solution was not retained, as items about food choices loaded on two different factors (i.e., items that were related to the same trait loaded on two different factors). The three factor solution did not include any items about diet or food choices. This was problematic because food choices are an essential component in global wellness.

The four factor solution was retained, as it was the only solution with distinct factors, which included items across all components of global wellness. The four-factor solution was also consistent with the 4-dimensional empirical framework (Kalkbrenner & Gormley, 2020) from which the LPHCI-2 items were developed. Consistent with the Aerobic Exercise LPHCI scale, the nine items that loaded on the first factor were related to physical activity (example items: “moved with the intention of getting physical activity” and “engaged in a physical activity that made you to sweat”). Eight items comprised the second factor, which was similar to the Consciousness of Stress LPHCI scale (example items: “had little energy, or pleasure/interest in doing things” and “felt down, depressed or hopeless”). The nine items that comprised the third factor resembled the Self-Care LPHCI scale (example items: “Been satisfied with your quality of life” and “felt close to another person”). Finally, the fourth factor included three items and was similar to the Food Choices LPHCI scale (example item: “Eaten a meal that included fresh fruits and/or vegetables”).

Rasch Item Response Theory Analyses

EFA is based on classical test theory (CTT), which is predicated on the assumption that each test item appraises the same degree (or difficulty; Stemler & Naples, 2021). Rasch item response theory (IRT) analyses extends CTT by uncovering the difficulty of test items and probability of a correct response. In other words, the EFA coefficients tend to indicate differing amounts of shared-variance represented by each item, however, CTT makes no inherent distinctions about the difficulty (or degree) of the latent trait appraised by each item. One of the major challenges in developing brief measures is reducing the number of items while maintaining the depth of the latent trait (Clark & Watson, 2019). Rasch IRT has utility for aiding in the development of brief screening tools, as the analysis identify items that are “maximally informative” (p. 1420). There is much debate in the measurement literature regarding whether Rasch measurement is a special case of IRT or if Rasch measurement and IRT are fundamentally different (Aryadoust et al., 2021; Stemler & Naples, 2021). Summarizing this debate is beyond the scope of this manuscript, however, most psychometric researchers agree that a fundamental assumption of both Rasch measurement and IRT is the ability reflected in test takers’ scores is a function of the latent variable that the test was designed to measure.

A Rasch IRT polytomous analysis was conducted in Jamovi version 2.2.5. A 1-PL Rasch model was tested, as 2-PL IRT and 3-PL IRT models have a tendency to increase the chances of problems with fit assessment, parameter estimation, and interpretability of results with newly developed measures (Fischer & Molenaar, 1995). For each of the four factors that emerged in the EFA, we computed Q3 coefficients to ensure the assumption of local independence with a cutoff value > −/+ .30 (Aryadoust et al., 2021). Infit MnSq and outfit MnSq item statistics were computed, with an acceptable range of 0.6 and 1.40 (Bond & Fox, 2007). Wright maps and expected score curves were computed for each item. Results revealed acceptable Person reliability coefficients for all four factors, .89, .75, .85, and 73, respectively. Six items were removed based on the Rasch analysis for violating the assumption of local independence or due to exceeding the acceptable infit/outfit range.

Research Team Item Reduction

There are no absolute guidelines in the literature for the length of short forms, however, 10 items or fewer are common for short forms in psychological research (Shields et al., 2021). While the EFA and Rasch IRT analysis reduced the original pool of items substantially, 23 items still remained following these analyses. The decision for retaining test items should be made based on both empirical (statistical) and theoretical (logical) bases (American Educational Research Association., American Psychological Association., National Council on Measurement in Education., & Joint Committee on Standards for Educational and Psychological Testing [AERA/APA/NCME/JCSEPT], 2014; Clark & Watson, 2019). Accordingly, the research team engaged in a consensus item evaluation and reduction process (Kalkbrenner, 2021) to ensure that the brief measure preserved the content coverage of the original LPHCI scales. Specifically, research team members first individually evaluated the remaining items for conceptual similarity with the empirical framework and theoretical blueprint with the goal of reducing the item pool to 10 or fewer, while maintaining the content coverage of the original LPHCI scales.

The research team came together for a consensus meeting and took turns presenting their individual work (i.e., their rationale for item removal based on the empirical framework and blueprint). The team first identified items with overlapping content, for example, “eaten a meal that included fresh fruits and/or vegetables” and “eaten a piece of fresh or organic fruit or vegetable.” The team then moved onto selecting the clearest items that reflected the four dimensions of the original LPHCI. Finally, researchers modeled the approximate proportion of LPHCI-Brief items based on the content of the original scale; including approximately twice as many mental wellness items (e.g., “felt hopeful about the future” and “felt proud of yourself”) than physical health items (e.g., “engaged in physical activity or exercise”). The research team eventually reached a consensus on the final pool of eight LPHCI-2: Brief Version items. The same three expert reviewers (see the “Instrument Development Process” section above) examined and approved of the content validity for the eight items.

Exploratory Factor Analysis with Reduced Item Pool

The EFA was recomputed to uncover the dimensionality of the final pool of eight LPHCI-2: Brief Version items. The same EFA procedures were replicated, including preliminary assumption checking, factor extraction, rotation, and factor retention criteria (see the “Exploratory Factor Analysis” section above). Retainable 1-factor and 2-factor solutions emerged (see Table 1). Both the one and two-factor solutions were retained, as it is not uncommon for EFA to reveal multiple potentially viable factor models. The 1-factor and 2-factor solutions were also consistent with the theoretical premise of total-wellness (i.e., mental and physical health are separate dimensions of a related latent construct).

Table 1.

Exploratory Factor Analysis Results: LPHCI-Brief One and Two-Factor Solutions

One Factor Solution Two Factor Solution
Item Content Factor Loading h2 Mental Wellness Physical Wellness h2
5. Been satisfied with your quality of life .74 .55 .84 .64
1. Felt proud of yourself .77 .58 .74 .61
8. Felt hopeful about the future .64 .40 .73 .47
6. Engaged in an activity that you enjoy .74 .54 .53 .53
3. Felt close to another person .68 .46 .50 .45
4. Eaten a meal that included fresh fruits and/or vegetables .51 .29 .74 .48
7. Ate meals/snacks on time .48 .23 .51 .31
2. Engaged in physical activity or exercise .51 .30 .47 .32
Open in a new tab

Note. Factor loadings that mark each particular factor appear in bold. Blank cells depict factor loadings < .10.

For the two-factor solution, all of the items that marked Factor 1 were related to mental health (see Table 1). Accordingly, Factor 1 was named Mental Wellness. Collectively, the items that comprised Factor 2 represented physical health in terms of diet and exercise. Factor 2 was named Physical Wellness. In terms of the one-factor solution, we looked closely at items 4 and 7 (see Table 1), as these items displayed h2 estimates below .30. The research team decided to keep these items for the time being primarily due to the theoretical relevance of these items. (AERA/APA/NCME/JCSEPT, 2014; Clark & Watson, 2019). Specifically, the content of items 4 and 7 (healthy diet; see Table 1) are essential components of global wellness. In addition, the h2 estimate for item 4 (.29) was very close to .30. Finally, both the one and two-factor solutions were about to be tested via CFA, which will reveal potential dimensionality issues more precisely, as CFA is a much more stringent test of internal structure than EFA. Internal consistency reliability estimates for scores on the single factor model (α = .84, ω = .84, H = .87) were all in the acceptable range (Nájera Catalán, 2019; Nunnally & Bernstein, 1994; Tavakol & Dennick, 2011).

The two factor solution demonstrated simple structure (see Table 1). Internal consistency reliability estimates were all in the acceptable range for Factor 1 (α = .84, ω = .84, H = .85). The internal consistency reliability estimates were in the acceptable-to-questionable range for Factor 2 (α = .62., ω = .62, H = .63). Internal consistency reliability estimates of test scores tend to be lower for scales with fewer items (Clark & Watson, 2019). The research team elected to proceed with the two-factor solution for the following reasons, (a) Factor 2 is a short scale (three items), (b) the standard error of measurement, as the omega estimate was very close to the recommended threshold (ω ≥ .65; Nájera Catalán, 2019), (c) the Physical Wellness scale is consistent with the empirical framework and is an essential component of measuring total wellness, and (d) there will be another opportunity to check the reliability of scores on the Physical Wellness scale with the second, CFA sample.

Confirmatory Factor Analysis

The eight LPHCI-2: Brief Version items that marked the one and two factor solutions (see Figure 1, Models 1 and 2) were entered into two separate CFAs. The CFAs were computed in IBM SPSS AMOS version 26 with maximum likelihood estimation methods. Collectively, the fit indices displayed an acceptable model fit for the one-factor solution (see Figure 1, Model 1): CMIN, χ2 (20) = 65.13, p < .001, CFI = .95, TLI = .93, RMSEA = .09, 90% CI (.06, .11), and SRMR = .05. The RMSEA estimate was slightly outside of the recommended range, however, the collective package of fit indices demonstrated acceptable internal structure. The standardized factor loadings for the one-factor solution are depicted in Figure 1 and ranged from .52 to .79. Internal consistency reliability estimates were in the acceptable range for scores on the one-factor model (α = .84, ω = .86, H = .88).

Figure 1.

Figure 1

Open in a new tab

Three Potential Models of the Structure of the LPHCI-2: Brief Version

The two-factor solution (see Figure 1, Model 2) displayed a very good model fit: CMIN, χ2 (19) = 35.29, p = .013, CFI = .98; TLI = .97, RMSEA = .05, 90% CI (.02, .08); and SRMR = .03. The standardized factor loadings are depicted in Figure 1 and ranged from .59 to .80. Internal consistency reliability estimates were in the acceptable range for scores on Mental Wellness (α = .84, ω = .85, H = .86). For Physical Wellness, the coefficient omega (ω = .70) and H (H = .70) estimates were in the acceptable range and the coefficient alpha estimate (α = .69) was in the acceptable-to-questionable range. We concluded acceptable internal consistency reliability estimates of scores on the Physical Wellness scale based on the omega and H estimates. Since both the unidimensional and two-dimensional LPHCI-Brief models displayed acceptable fit, a chi-squared test of difference was computed to test for superiority of fit between the two models. Results revealed that the two-dimensional model (see Figure 1, Model 2) demonstrated a significantly superior fit with the data compared to the unidimensional model (see Figure 1, Model 1), X 2(1) = 29.84, p < .001.

The estimate of the co-variance between factors for the two-factor solution (see Figure 1, Model 2) was in the strong range, which suggests that the Mental Wellness and Physical Wellness subscales might be appraising the same construct on a statistical level. However, strong correlations between single-order factors can also suggest a higher-order latent trait (Credé & Harms, 2015). In addition to strong correlations between factors, theoretical support, and a poorly fitting unidimensional model suggest that a higher-order factor might be present in the data. Theoretical support exists for a higher-order Global Wellness scale based on the findings of Kalkbrenner and Gormley (2020) and Kalkbrenner (2022b). The unidimensional model (Figure 1, Model 1) did not display poor fit, however, it did not fit as well as the two-dimensional model. Accordingly, a higher-order CFA was computed to test for a second order factor (see Figure 1, Model 3). Consistent with the two-dimensional model, the higher-factor solution (see Figure 1, Model 3) displayed a very good model fit: CMIN, χ2 (19) = 35.29, p = .013; CFI = .98; TLI = .97, RMSEA = .05, 90% CI (.02, .08); and SRMR = .03. The final eight LPHCI-2: Brief Version items (Kalkbrenner et al., 2023) are available in Appendix A.

Convergent Validity Testing

A Pearson product moment correlation was calculated to test the convergent validity between scores on the Global Wellness scales of the LPHCI-2: Brief Version and the original version of the LPHCI. A high correlation signifies convergent validity evidence of scores and according to Drummond et al. (2016), a correlation coefficient > .50 denotes a high correlation. Results supported convergent validity evidence by revealing a strong correlation, r = .61, r2 =.37, p < .001, 2-tailed, between scores on the Global Wellness scales of the LPHCI-2: Brief Version and the original version of the LPHCI.

Discussion

Counselors need brief screening tools with validated scores for measuring total wellness (mental and physical) considering the wellness orientation of the discipline (Myers, 1992), coupled with the growing prevalence and complexity of mental health issues and non-communicable diseases (CDC, 2022; NAMI, 2022). The aim of the present study was to develop and validate scores on the LPHCI-2: Brief Version. The research team followed rigorous and empirically supported steps (MEASURE approach) throughout the item development and score validation process. Consistent with established brief measures in the literature (e.g., PHQ-9 and GAD-7), the results of internal structure validity testing (EFA and CFA) in the present study revealed and confirmed the dimensionality of both a two-factor and higher-order factor solution for the LPHCI-2: Brief Version. In an extension of the extant literature, the LPHCI-2: Brief Version offers researchers and practitioners a brief screening tool for simultaneously appraising mental and physical wellness.

Collectively, tests of internal structure met or exceeded the recommendations of leading psychometric researchers for acceptable internal structure of scores on both the unidimensional and two-dimensional models (Dimitrov, 2012; Schreiber et al., 2006). The CFA results revealed stronger support for the two-dimensional (Figure 1, Model 2) and higher order (Figure 1, Model 3) LPHCI-brief model than the unidimensional model (Figure 1, Model 1). The two-factor and higher-order models offer three potentially valuable scales for use in clinical and research settings. The Mental Wellness and Physical Wellness scales that comprise two-factor LPHCI brief model (Figure 1, Model 2) can be used in instances where one is aiming to parse out respondents’ state of mental and/or physical health. The higher-order factor model (Figure 1, Model 3) might have utility in situations where researchers or practitioners are seeking to measure respondents’ collective mental and physical health (global wellness).

Brief scales tend to measure their intended latent traits on a more general level. The strong positive correlation that emerged between scores on the original LPHCI and the LPHCI-2: Brief Version supported convergent validity evidence. The strength of the correlation (strong, however, not excessive) was in the ideal range for scores on a new version of a screening tool. In fact, our decision to create a new version for the LPHCI brief (rather than a reduced pool of the original LPHCI items) was due to the specific content of the original LPHCI items. Accordingly, the results of convergent validity suggest that the LPHCI-2: Brief Version is probably measuring a slightly more general form of global wellness than the original version, which was to be expected since the brief version is comprised of different and more general items than the original.

Limitations and Future Research

The results and implications of the preset study should be considered in the context of the methodological limitations. Our findings supported construct validity evidence of scores on the LPHCI-2: Brief Version among a national sample of adults in the United States. However, the sample was not necessarily a nationally representative sample. Relatedly, the cross-sectional design limits the temporal implications of the results. Future research is needed to test how and in what ways global wellness might fluctuate over time. Specifically, future researchers can test the capacity of LPHCI-2 scores for predicting wellness-related outcome variables. In addition, global wellness can vary by demographic variables (Kalkbrenner, 2022a). Accordingly, future researchers should test the factorial invariance of LPHCI-2: Brief Version scores among of adults in the U.S. by key demographic variables that are associated with total wellness. For example, future researchers can test the psychometric properties of the LPHCI-2: Brief Version with/by college students, SES, age, gender, older adults, and ethnic identity.

The EFA revealed a four-factor solution that was consistent with the original LPHCI. However, there were far too many items in the final four-factor solution to comprise a short form. Accordingly, we used Rasch IRT and a research team item reduction process to identify the most informative items for the short form and then recomputed the EFA to uncover the dimensionality of the revised pool of items. The second EFA revealed a different factor solution (two factors instead of four factors). However, it is possible that we capitalized on chance since the EFAs and Rasch IRT analysis were computed with the first sample.

The majority of the reliability estimates for scores on the Global Wellness and Mental Wellness scales were consistently in the acceptable range in both the EFA and CFA samples. However, the internal consistency reliability estimates for the Physical Wellness scale were in the questionable range for the EFA sample. In the CFA sample, α was just below (α = .69) the recommended threshold (α ≥ .70), however, ω and H were in the acceptable range. It is promising that the ω and H estimates were satisfactory in the CFA sample, as composite reliability estimates are more stable than α (McNeish, 2018). However, the questionable reliability evidence for scores on the Physical Wellness scale in the EFA sample should not be ignored. It is possible that the few number of items (N = 3) that comprise the Physical Wellness scale accounted for the questionable reliability estimates of scores, as reliability coefficients tend to be lower for shorter scales (Clark & Watson, 2019). Future researchers should carefully test the internal consistency reliability of scores on the Physical Wellness scale (i.e., compute α, ω, and H) to ensure that the scale’s use is appropriate with their sample.

Implications for Practice

There is a growing need for brief measures with valid scores in both clinical and research settings (McCauley et al., 2021; Schipolowski et al., 2014; Shields et al., 2021). This growing need is driven by efficiency considerations (Lang & Connell, 2017; Marsh et al., 2020; Schipolowski et al., 2014), time constraints in IPC settings (Kleiman et al., 2020), and striving to reduce participant fatigue throughout this process (McDermott et al., 2019). The LPHCI-2: Brief Version has potential to edify research and practice in IPC settings by providing professional counselors and other IPC practitioners with a brief (< 10 items), free to use, electronically or paper copy available, and easy to score screening tool for appraising the key outcome variable in IPC settings (total wellness). The LPHCI-2: Brief Version is available for free and can be accessed by contacting the corresponding author.

The latent trait of global wellness is comprised of lifestyle practices and health consciousness, which are statistically significant predictors of mental and physical health (Kalkbrenner, 2022a; Ohrnberger et al., 2017). At the current stage of development, the LPHCI-2 does not include criterion-referenced cutoff scores and future research is needed to test the capacity of the LPHCI-2: Brief Version for measuring global wellness over time. However, counselors and other IPC practitioners can use the LPHCI-2: Brief Version to measure their clients’ global wellness at a distinct period of time. Counselors can score and interpret the higher-order LPHCI-2: Brief Version as one way to measure their clients’ general mental and physical health, which is a key outcome variable in IPC settings. The two-dimensional model (Mental Wellness and Physical Wellness subscales) might have utility for highlighting particularly healthy or unhealthy lifestyle practices for a specific client. Suppose, for example, a client scores higher on the Mental Wellness scale and lower on the Physical Wellness scale. The client and counselor can review their LPHCI-2: Brief Version scores and use the item content as one way to discuss how the client can sustain their positive lifestyle practices for mental wellness and engage in new lifestyle practices to improve their physical health.

Conclusion

There is a mounting necessity for brief measures with valid scores for measuring total wellness in clinical, research, and IPC settings. The purpose of the present study was to develop and validate scores on the LPHCI-2: Brief Version with a national sample of adults in the United States. The results of EFA, CFA, higher-order CFA, convergent validity, and internal consistency reliability testing revealed support for a two-dimensional (mental and physical wellness) and a higher-order model (global wellness) for the LPHCI brief version. The LPHCI-2: Brief Version offers counselors and other IPC practitioners a brief screening tool for measuring their clients’ global wellness, which can be used to establish baseline health levels, aid in treatment planning, and to monitor progress toward meeting the goals of treatment plans.

Funding:

This research was supported by an Institutional Development Award (IDeA) from the National Institute of General Medical Sciences of the National Institutes of Health under grant number P20GM103451.

Appendix A

The Lifestyle Practices and Health Consciousness Inventory-2: Brief Version

The following questions are about some aspects of your general lifestyle. Please respond to the questions using the following prompt and scale:

0 = never 1 = one to five times 2 = six to ten times 3 = eleven to fifteen times

4 = sixteen to twenty times 5 = twenty one or more times

In the past 30 days, how often have you….

  1. … Felt proud of yourself

  2. … Engaged in physical activity or exercise

  3. … Felt close to another person

  4. … Eaten a meal that included fresh fruits and/or vegetables

  5. … Been satisfied with your quality of life

  6. … Engaged in an activity that you enjoy

  7. … Ate meals/snacks on time

  8. … Felt hopeful about the future

Scoring Instructions

Global Wellness: Compute the average score across all 8 items.

Mental Wellness: Compute the average score for the following items: 1, 3, 5, 6, and 8.

Physical Wellness: Compute the average score for the following items: 2, 4, and 7.

Permission to use and reproduce The LPHCI-2: Brief Version is granted by the authors, provided (1) the authors are referenced appropriately, (2) no changes are made, and (3) it is not sold.

Footnotes

Conflict of Interest:

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

References

  1. American Counseling Association. (2014). ACA code of ethics. https://www.counseling.org/resources/aca-code-of-ethics.pdf
  2. American Educational Research Association., American Psychological Association., National Council on Measurement in Education., & Joint Committee on Standards for Educational and Psychological Testing. (2014). Standards for educational and psychological testing. https://www.apa.org/science/programs/testing/standards
  3. Aryadoust V, Ng LY, & Sayama H (2021). A comprehensive review of Rasch measurement in language assessment: Recommendations and guidelines for research. Language Testing, 38(1), 6–40. 10.1177/0265532220927487 [DOI] [Google Scholar]
  4. Beavers AA, Lounsbury JW, Richards JK, Huck SW, Skolits GJ, & Esquivel SL (2013). Practical considerations for using exploratory factor analysis in educational research. Practical Assessment, Research & Evaluation, 18(5/6), 1–13. https://scholarworks.umass.edu/pare/vol18/iss1/6/ [Google Scholar]
  5. Berwick DM, Murphy JM, Goldman PA, Ware JE Jr., Barsky AJ, Weinstein MC, Berwick DM, Murphy JM, Goldman PA, Ware JE Jr, Barsky AJ, & Weinstein MC (1991). Performance of a five-item mental health screening test. Medical Care, 29(2), 169–176. 10.1097/00005650-199102000-00008 [DOI] [PubMed] [Google Scholar]
  6. Bond TG, & Fox CM (2007). Applying the Rasch model: Fundamental measurement in the human sciences (2nd ed.). Lawrence Erlbaum. [Google Scholar]
  7. Browne RH (1995). On the use of a pilot sample for sample size determination. Statistics in Medicine, 14, 1933 – 1940. 10.1002/sim.4780141709 [DOI] [PubMed] [Google Scholar]
  8. Clark LE, & Watson D (2019). Constructing validity: New developments in creating objective measuring instruments. Psychological Assessment, 31(12), 1412–1427. 10.1037/pas0000626 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Credé M, & Harms P (2015). 25 years of higher-order confirmatory factor analysis in the organizational sciences: A critical review and development of reporting recommendations. Journal of Organizational Behavior, 36(6), 845–872. 10.1002/job.2008 [DOI] [Google Scholar]
  10. Dimitrov D (2012). Statistical methods for validation of assessment scale data in counseling and related fields. American Counseling Association. [Google Scholar]
  11. Drummond RJ, Sheperis CJ, & Jones KD (2016). Assessment procedures for counselors and helping professionals (8th ed.). Pearson [Google Scholar]
  12. Field AP (2018). Discovering statistics using IBM SPSS Statistics (5th ed.). Sage. [Google Scholar]
  13. Fischer GH, & Molenaar IW (1995). Rasch models: Foundations, recent developments, and applications. Springer-Verlag [Google Scholar]
  14. Hertzog M (2008). Considerations in determining sample size for pilot studies. Research in Nursing & Health, 31(2), 180–191. 10.1002/nur.20247 [DOI] [PubMed] [Google Scholar]
  15. Jimenez-Lara M (2016). Reaping the benefits of integrated health care. Stanford social innovation review: Informing and inspiring leaders of social change. https://ssir.org/articles/entry/reaping_the_benefits_of_integrated_health_care [Google Scholar]
  16. Kalkbrenner MT (2021). A practical guide to instrument development and score validation in the social sciences: The MEASURE Approach. Practical Assessment, Research & Evaluation. 26(1), Article 1. 10.7275/svg4-e671 [DOI] [Google Scholar]
  17. Kalkbrenner MT, Bradley M, & Sun H (2023). The Lifestyle Practices and Health Consciousness Inventory-2: Brief Version. [Screening Tool]. [DOI] [PMC free article] [PubMed]
  18. Kalkbrenner MT, & Gormley B (2020). Development and initial validation of scores on the Lifestyle Practices and Health Consciousness Inventory (LPHCI). Measurement and Evaluation in Counseling and Development. 53(4), 219–237. 10.1080/07481756.2020.1722703 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kalkbrenner MT (2022a). Global Wellness: Predicting lower levels of anxiety and depression severity. Journal of Counseling and Development. 100(1), 84–95. 10.1002/jcad.12405 [DOI] [Google Scholar]
  20. Kalkbrenner MT (2022b). Validation of scores on The Lifestyle Practices and Health Consciousness Inventory with Black and Latinx adults in the United States: A three-dimensional model. Measurement and Evaluation in Counseling and Development. 55(2), 84 – 97. 10.1080/13607863.2018.1544217 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kleiman SE, Bovin MJ, Black SK, Rodriguez P, Brown LG, Brown ME, Lunney CA, Weathers FW, Schnurr PP, Spira J, Keane TM, & Marx BP (2020). Psychometric properties of a brief measure of posttraumatic stress disorder–related impairment: The Brief Inventory of Psychosocial Functioning. Psychological Services, 17(2), 187–194. 10.1037/ser0000306 [DOI] [PubMed] [Google Scholar]
  22. Kroenke K, Spitzer R, & Williams J (2001). The PHQ- 9. Journal of General Internal Medicine, 16(9), 606–613. 10.1046/j.1525-1497.2001.016009606.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lang JM, & Connell CM (2017). Development and validation of a brief trauma screening measure for children: The Child Trauma Screen. Psychological Trauma: Theory, Research, Practice, and Policy, 9(3), 390–398. 10.1037/pas0000787 [DOI] [PubMed] [Google Scholar]
  24. Marsh HW, Huppert FA, Donald JN, Horwood MS, & Sahdra BK (2020). The well-being profile (WB-Pro): Creating a theoretically based multidimensional measure of well-being to advance theory, research, policy, and practice. Psychological Assessment, 32(3), 294–313. 10.1037/pas0000787 [DOI] [PubMed] [Google Scholar]
  25. McCauley L, Phillips RL, & Robinson SK (2021). Implementing high-quality primary care: Rebuilding the foundation of healthcare. A consensus study report of the National Academies of Sciences, Engineering, Medicine. National Academic Press. [Google Scholar]
  26. McDermott RC, Hammer JH, Levant RF, Borgogna NC, & McKelvey DK (2019). Development and validation of a five-item male role norms inventory using bifactor modeling. Psychology of Men & Masculinities, 20(4), 467–477. 10.1037/men0000178 [DOI] [Google Scholar]
  27. McNeish D (2018). Thanks coefficient alpha, we’ll take it from here. Psychological Methods, 23(3), 412–433. 10.1037/met0000144 [DOI] [PubMed] [Google Scholar]
  28. Mulvaney-Day N, Marshall T, Downey Piscopo K, Korsen N, Lynch S, Karnell LH, Moran GE, Daniels AS, & Ghose SS (2018). Screening for behavioral health conditions in primary care settings: A systematic review of the literature. Journal of General Internal Medicine, 33(3), 335–346. 10.1007/s11606-017-4181-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Mvududu NH, & Sink CA (2013). Factor analysis in counseling research and practice. Counseling Outcome Research and Evaluation, 4(2), 75–98. 10.1177/2150137813494766 [DOI] [Google Scholar]
  30. Myers JE (1992). Wellness, prevention, development: The cornerstone of the profession. Journal of Counseling & Development, 71(2), 136–139. 10.1002/j.1556-6676.1992.tb02188.x [DOI] [Google Scholar]
  31. Myers JE, & Sweeney TJ (2014). The Five Factor Wellness Inventory, Adult (5F-Wel-A). Mind Garden. [Google Scholar]
  32. Nájera Catalán H (2019). Reliability, population classification and weighting in multidimensional poverty measurement: A Monte Carlo study. Social Indicators Research, 142(3), 887–910. 10.1007/s11205-018-1950-z [DOI] [Google Scholar]
  33. National Alliance on Mental Illness. (2022, February). Mental health by the numbers. https://www.nami.org/mhstats
  34. Center for Disease Control and Prevention. (2022). Heart Disease Facts. https://www.cdc.gov/heartdisease/facts.htm#:~:text=Heart%20disease%20is%20the%20leading,groups%20in%20the%20United%20States.&text=One%20person%20dies%20every%2034,United%20States%20from%20cardiovascular%20disease.&text=About%20697%2C000%20people%20in%20the,1%20in%20every%205%20deaths.
  35. Nunnally JC & Bernstein IH (1994). Psychometric Theory (3rd ed.). McGraw Hill. [Google Scholar]
  36. Ohrnberger J Fichera E, & Sutton M (2017). The relationship between physical and mental health: A mediation analysis. Social Science & Medicine, 195(1), 42–49. 10.1016/j.socscimed.2017.11.008 [DOI] [PubMed] [Google Scholar]
  37. Qualtrics Sample Services [Online sampling service service]. (2022). Unlock breakthrough insights with market research panels. https://www.qualtrics.com/research-services/online-sample/
  38. Schipolowski S, Schroeders U, & Wilhelm O (2014). Pitfalls and challenges in constructing short forms of cognitive ability measures. Journal of Individual Differences, 35(4), 190–200. 10.1027/1614-0001/a000134 [DOI] [Google Scholar]
  39. Schreiber JB, Nora A, Stage FK, Barlow EA, & King J (2006). Reporting structural equation modeling and confirmatory factor analysis results: A review. Journal of Educational Research, 99(6), 323–337. 10.3200/JOER.99.6.323-338 [DOI] [Google Scholar]
  40. Shannonhouse L, Erford B, Gibson D, O’Hara C, & Fullen M (2020). Psychometric synthesis of the Five Factor Wellness Inventory. Journal of Counseling & Development, 98(1), 94–106. 10.1002/jcad.12303 [DOI] [Google Scholar]
  41. Shields RE, Korol S, Carleton RN, McElheran M, Stelnicki AM, Groll D, & Anderson GS (2021). Brief mental health disorder screening questionnaires and use with public safety personnel: A review. International Journal of Environmental Research and Public Health, 18(7), 1–30. 10.3390/ijerph18073743 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Spitzer RL, Kroenke K, Williams JBW, & Löwe B (2006). A brief measure for assessing Generalized Anxiety Disorder: The GAD-7. Archives of Internal Medicine, 166(10), 1092–1097. 10.1001/archinte.166.10.1092 [DOI] [PubMed] [Google Scholar]
  43. Stemler SE, & Naples A (2021). Rasch Measurement v. Item Response Theory: Knowing when to cross the line. Practical Assessment, Research & Evaluation, 26(11), 1–16. https://scholarworks.umass.edu/pare/vol26/iss1/11/ [Google Scholar]
  44. Tavakol M, & Dennick R (2011). Making sense of Cronbach’s alpha. International Journal of Medical Education, 2(1), 53–55. 10.5116/ijme.4dfb.8dfd [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. United States Census Bureau. (2020). Quick facts: United States. https://www.census.gov/quickfacts/fact/table/US/PST045222

RESOURCES