This page provides answers to some of the frequently asked questions about the IBM-HS and help-seeking research in general. More answers will be added over time, so check back periodically. If you have a question not answered on this page, please consider reviewing the relevant sections of this website for potential answers before inquiring directly with Dr. Hammer.
Q: Do I need to measure help-seeking beliefs if I am going to use the IBM-HS?
A: No. Help-seeking beliefs, which are typically measured by indirect belief measures, provide value to help-seeking investigations. However, because indirect measures are population-specific and therefore take time and resources to develop (for more information on developing indirect belief measures, see the mixed-method protocol page), it is not always feasible to assess help-seeking beliefs with indirect measures. And, for some help-seeking investigations, help-seeking beliefs are not relevant to the questions being investigated. Help-seeking belief measures are a nice way to efficiently assess a wide range of topic-specific beliefs about mental health help seeking, so that beliefs/aspects that matter the most (i.e., correlate strongly with their corresponding help-seeking mechanism and/or with help-seeking intention) can be distinguished from those beliefs/aspects that don’t matter for the population in question, which is very useful when you have limited time/resources and need to make difficult decisions about which beliefs/aspects to prioritize targeting via future interventions designed to increase help-seeking behavior in the population of interest.
Q: Does the IBM-HS consider [name of construct] to be a help-seeking determinant, help-seeking belief, help-seeking mechanism?
A: It is a common aphorism in social science that “all models are wrong, but some are useful.” The IBM-HS, like the reasoned action theoretical tradition from which it was developed, is a useful model that necessarily oversimplifies the complexity of the mental health help-seeking process. The IBM-HS packages clusters of constructs in a manner that offers some conceptual advantages to professionals seeking to achieve an accessible understanding of what shapes the help-seeking process. However, this packaging also introduces some downsides. One limitation is that it can be hard to cleanly link certain bodies of social science/medical research literature to the IBM-HS’s construct taxonomy. For example, the literature on stigma, specifically the stigma of mental illness and the stigma of seeking mental healthcare, discusses a variety of types of stigma. I recommend Brenner and colleagues’ (2022) chapter on disentangling mental illness and help-seeking stigmas (Publisher Version; Pre-Press Version) for a helpful overview. These different types of stigma do not all fall into one construct category within the IBM-HS. Stigma is a multidimensional phenomenon that does not correspond to a single construct in the IBM-HS. Depending on how a stigma-related construct is conceptualized and measured, it may represent a help-seeking determinant, belief, or mechanism. Because existing stigma measures sometimes encompass multiple conceptual domains, researchers should examine individual item content rather than assume that a measure’s traditional name identifies a single IBM-HS construct. Here’s a crosswalk chart where Dr. Hammer tried his best to link types of stigma with the IBM-HS construct taxonomy:
| Stigma construct or perception | IBM-HS construct category | Specific IBM-HS construct | Notes / Relevant Literature |
| Self-stigma of mental illness: Experiencing or anticipating reduced self-worth because one has a mental illness (e.g., SSOMI) | Determinant | Mental health perceptions, knowledge, and skills | It’s not an outcome belief because the IBM-HS outcome beliefs are focused on the perceived attributes/outcomes of seeking professional mental health care, not the attributes or outcomes of having mental illness. |
| Social-network stigma of mental illness: Believing that friends, family, or other close individuals would think less of oneself if they knew one had a mental illness | Determinant | Structural forces (perception of stigma) | It’s not an outcome belief because the IBM-HS outcome beliefs are focused on the perceived attributes/outcomes of seeking professional mental health care, not the attributes or outcomes of having mental illness. |
| Public stigma of mental illness: Societal stigma held by members of the general public towards those who have or are perceived to have mental illness (e.g., PDD) | Determinant | Structural forces (perception of stigma) | Perceptions of prejudice/discrimination on the basis of a stigmatized identity are categorized as structural forces by the IBM-HS. |
| Personal stigma of mental illness: Respondent’s own endorsement of stigmatizing stereotypes, prejudicial attitudes, or discriminatory inclinations toward people with mental illness | Determinant | Mental health perceptions, knowledge, and skills (though a case can be made for Structural Forces as well) | Schnyder et al., (2017); Fox et al., (2018); Griffiths et al., (2004); Griffiths et al., (2008) Holding a prejudicial belief toward people with a given stigmatized identity is a mental health-related perception that is driven by the structural force of ableism. |
| Self-stigma of seeking help: Believing one would think less of oneself if one sought professional mental health care (e.g. SSOSH) | Belief | Outcome belief | Perceiving that one would think less of oneself as as result of seeking help is an anticipated outcome. |
| Social-network stigma of seeking help: Believing that people in one’s social network would think less of oneself if one sought professional mental health care (e.g., PSOSH) | Belief | Outcome belief | Perceived that close others would think less of oneself as a result of seeking help is an anticipated outcome. It is similar to, but independent from, the perception of whether the people around you think you should or should not seek help (i.e., injunctive perceived norm). |
| Public stigma of seeking help: Societal stigma held by members of the general public towards those who have or are perceived to have sought professional mental health care (e.g., SSRPH) | Determinant | Structural forces (perception of stigma) | Perceptions of prejudice/discrimination on the basis of a stigmatized identity are categorized as structural forces by the IBM-HS. It is related to, but independent from, the perception of whether most people think you should or should not seek help (i.e., injunctive perceived norm). |
| Personal stigma of seeking help: Respondent’s own endorsement of stigmatizing stereotypes, prejudicial attitudes, or discriminatory inclinations toward people who have sought professional mental health care | Determinant | Mental health perceptions, knowledge, and skills (though a case can be made for Structural Forces as well) | Holding a prejudicial belief toward people with a given stigmatized identity is a mental health-related perception that is driven by the structural force of ableism. |
| Interdependent stigma of seeking help: the extent to which people believe their family would be devalued and shamed if they seek psychological help (e.g., ISSH) | Belief | Outcome belief | Vogel et al., (2024) |
| Stigma-related shame or embarrassment of seeking help: Experiencing shame or embarrassment when contemplating the prospect of seeking help | Belief | Experiential belief | Experiential beliefs are focused on the raw feelings/affect that arises in the moment when a person contemplates the idea of their potentially seeking professional help, rather than the cognitive reasons that lead to that feeling. |
| Perceived disapproval of seeking help by particular referents: Believing that one’s spouse, friends, family, or other specific referents would disapprove of one’s seeking help | Belief | Beliefs about others’ expectations | Specific close others who negatively judge people for seeking help [i.e., social network stigma] would be likely to also disapprove of the respondent seeking help [i.e., beliefs about others’ expectations, that subsequently shape injunctive perceived norm]. They are two strongly correlated constructs, rather than being the same construct. |
| Overall perceived social disapproval of seeking help: Perceiving that, collectively, the important people in one’s life would disapprove of one’s seeking help | Mechanism | Perceived Norm: Injunctive | Close others as a general group who negatively judge people for seeking help [i.e., social network stigma] would be likely to also collectively disapprove of the respondent seeking help [i.e., injunctive perceived norm]. They are two strongly correlated constructs, rather than being the same construct. |
| Believing that particular referents would avoid seeking help themselves | Belief | Beliefs about others’ behavior | |
| Overall perception that important others avoid seeking help: Believing that most important people in one’s life would not seek professional help themselves | Mechanism | Perceived Norm: Descriptive |
Q: If I am not using the IBM-HS as the guiding theory for my investigation, is it OK with use some IBM-HS measures?
A: Yes, use whatever will be helpful to you. It is common for help-seeking researchers to mix and match across different theories and measures. Just be sure to consider the principle of compatibility when using any measures of beliefs, mechanisms, or intention derived from the reasoned action tradition (which includes the IBM-HS).
Q: Where can I find information about, get a copy of, or request permission to use/translate your measures?
A: All of my first-author measures have a webpage dedicated to providing all of this, so look up the measure(s) of interest to you by starting from the Measures webpage and navigating from there. If I am not the first/corresponding author of a given instrument, I cannot grant you permission to use that instrument, because it’s not my intellectual property. Please note that the ISMI-29/ISMI-10 are the intellectual property of Dr. Jennifer Boyd (https://www.researchgate.net/profile/Jennifer-Boyd-2/research) and I cannot grant permission on her behalf. I have been told by several researchers that Dr. Boyd has not responded to their emails so scholars may need to make the choice between not using the ISMI-29/10 at all or using it despite not having written permission from Dr. Boyd.
Q: Does [name of Dr. Hammer’s instrument] have cut-off scores, norms, categories, ranges, classifications?
A: In general, Dr. Hammer’s instruments provide only continuous scores. We do not have population norms or validated categorical cutoffs in most cases. Users may compare scores descriptively using the means, standard deviations, and percentiles from the validation sample, but “low/medium/high” categories should not be interpreted as established norms unless future norming studies are conducted in the relevant target population. It is very common for Dr. Hammer to receive emails from researchers asking “how do I know what score represents a low level of [construct] vs. a medium or high level?”. Dr. Hammer can only clarify that such range/categories/descriptors are completely dependent on what is normative in a given population. Therefore, unless past researchers have collected a representative sample of a given population in order to determine what scores are low vs. medium vs. high relative to the average member of that population, it is not appropriate to say something like “scores from X to Y indicate low [construct]”. For further explanation, see the “Can I put respondents into categories based on their MHSIS score?” section of the MHSIS measure webpage. However, it is OK to tentatively interpret a given score relative to the response scaling labels for that measure; for example, if the 3-item MHSIS uses a 6-point Likert scale, then it might be reasonable to say that a score of 1 or 2 could be considered a low or weak degree of intention, a score of 3 or 4 could be considered a moderate or mild degree of intention, and a score of 5 or 6 could be considered a high or strong degree of intention.
Q: Can I use single-item versions of your direct measures of the IBM-HS mechanisms and intention?
A: As of October 2026, we have a paper under review that seeks to provide evidence of validity for single-item versions of several of the IBM-HS direct measures. When that is published, that will provide a strong rationale for being able to use single-item versions of certain measures in certain situations. Until that paper is published, however, you have to be careful about using single-item versions–use them at your own risk. The unpublished findings we have so far indicate that intention is best measured with 1 item, attitude is best measured with 3 items, and the four perceived norm and personal agency facets are best measured with 1 item apiece. Thus, it’s likely that most researchers will be using shorter versions of these measures in the future, but the validity evidence for this has not been published yet. And just because single item measures work well with some populations or in traditional written United States English, does not mean it will work well in other populations or when translated into other languages; future psychometric work will need to investigate that.
Q: What are some good examples of peer-reviewed papers that use the IBM-HS in different ways?
A: Below are lists of papers that use different aspects of the IBM-HS but not others, so that you can see the ways to adapt use of the IBM-HS to your research needs and design choices.
- Papers that use the IBM-HS as a theoretical framework to guide qualitative data collection and analysis
- Ban, N., Mensah, L. O., Whitwer, M., Hargis, L. E., Wright, C. J., Hammer, J. H., & Wilson, S. A. (2023, June). “It’s very important to my professors…at least most of them”: How messages from engineering faculty and staff influence student beliefs around seeking help for their mental health. American Society for Engineering Education Annual Conference, Baltimore, MD. (PDF)
- Wright, C. J., Wilson, S. A., Hammer, J. H., Hargis, L. E., Miller, M. E.*, & Usher, E. L. (2023). Mental health in undergraduate engineering students: Identifying facilitators and barriers to seeking help. Journal of Engineering Education, 112(4), 963-986. https://doi.org/10.1002/jee.20551 (PDF) *Note: text formally cites the IBM because the IBM-HS paper was not yet published at the time.
- Papers that use the IBM-HS as a theoretical framework to guide selection of quantitative help-seeking measures, statistical model construction, and quantitative analysis:
- Kaiser, S.*, Hammer, J. H., Shepherd, L.*, & Weng, W.* (in press). Exploring Sociocultural Predictors of Intention to Seek Help Among Older Adults using the Integrated Behavioral Model of Mental Health Help Seeking. Journal of Counseling Psychology. https://psycnet.apa.org/doi/10.1037/cou0000895 (PDF forthcoming)
- Weng, W.*, Hammer, J. H., Kim, N.*, McDermott, R., Li, M., Wang, F., & Shi, X. (in press). Understanding Chinese college students’ professional help-seeking intention: A moderated integrated behavioral model. The Counseling Psychologist. https://doi.org/10.1177/00110000261458370 (PDF forthcoming).
- Whitwer, M., Hammer, J. H., Gomer, B., Vahidi, E., & Wilson, S. A. (2025). When performance takes priority: Beliefs that shape engineering students’ mental health help-seeking. Education Sciences, 15(11), 1553. https://doi.org/10.3390/educsci15111553 (PDF)
- Day, A.* & Hammer, J. H. (2025). Exploring sociocultural predictors of intention to seek help among South Asians in the United States using the Integrated Behavioral Model of Mental Health Help-Seeking. Asian American Journal of Psychology, 17(1), 71-83. https://doi.org/10.1037/aap0000391 (PDF forthcoming).
- Papers that use the IBM-HS as a theoretical framework to structure a literature review
- Park, D., Lee, S., & Lee, D. (2026). Research trends in professional mental health help-seeking among Korean older adults: A systematic review applying the Integrated Behavioral Model of Mental Health Help Seeking. Korean Journal of Gerontological Social Welfare, 81(3), 103–137. https://doi.org/10.21194/kjgsw.81.3.202609.103
- Funded grant proposals that used IBM-HS as a part of the theoretical framework
- Collaborative Research: Mental Health Care Pathways: Examining Help Seeking Behaviors in Engineering Graduate Students (2026-2031). National Science Foundation. https://www.nsf.gov/awardsearch/show-award?AWD_ID=2556259
- Research: Identifying intervention targets to increase mental health help seeking in undergraduate engineers (2022-2025). National Science Foundation. https://www.nsf.gov/awardsearch/show-award/?AWD_ID=2225567
Q: How does the IBM-HS think about the relationship between the concepts/constructs of perceived need; evaluated need; psychological distress; and mental health perceptions, knowledge, and skills?
A: See the Evaluated Need Measures and Mental Health Perceptions, Knowledge, and Skills Measures webpages for more information. Please note that the construct of “perceived need” is not formally modeled within the IBM-HS taxonomy as its own construct. Dr. Hammer thinks about perceived need as being shaped by a number of factors, including a person’s subjective perceptions of (1) their own mental health status/symptoms/functioning and (2) the nature, effectiveness, and utility of available professional mental health care treatment. A person has to believe they have a problem that needs fixing and that professional mental health treatment is an appropriate solution to that problem. If both are not true, then it’s unlikely a person will perceive a need for them to get professional treatment. Next, perceiving a need for treatment is, generally, a prerequisite for intending to seek professional treatment. After all, except in special cases of involuntary treatment or strong social pressure, why would a person intend to seek professional help if they didn’t think they needed professional help? Thus, perceived need could be considered a predictor/cause of intention, which has a complex relationship with the IBM-HS help-seeking mechanisms: some of these mechanisms may causally shape perceived need, have an indirect effect on intention via perceived need, and/or have a unique incremental direct effect on intention. Some of the other behavioral prediction theories put more weight on the concept of perceived need than does the IBM-HS.
Q: Is perceived likelihood of seeking professional help the same as help-seeking intention?
A: It depends on which scholar you ask, which paper you read, or which measure you use. Dr. Hammer is aware that published research studies often find that items that ask “how likely is it that you will seek help” (help-seeking likelihood) are strongly correlated with items that ask “to what degree do you intend to seek help” (help-seeking intention), sometimes so strongly that it would be hard to argue that the items are measuring different constructs (see this wonderful Lawson & Robins 2021 article on Sibling Constructs for a helpful way to think about when two constructs should or should not be considered independent). However, conceptually, Dr. Hammer would argue that intention and likelihood are different and should not be treated as they same (and thus, we should not use items assessing both to measure a single global concept called “help-seeking intention”). For example, you can fully intend to start eating healthy and going to the gym three times a week, but when asked how likely it is that will happen, to admit it’s not that likely. As another humorous example, consider the phrase “the road to hell is paved with good intentions“. Thus, Dr. Hammer believes that the cleanest and most valid way to measure help-seeking intention is to ask about one’s planned readiness to perform help-seeking behavior. There are certainly a number of factors (i.e., intention enactment factors) that can shape whether or not that intention ultimately translates into successful behavior, which then shapes the likelihood that one would seek help.
Q: What is the relationship between “willingness to seek help” and “intention to seek help”?
A: It depends on which scholar you ask, which paper you read, or which measure you use. Dr. Hammer’s personal preference has shifted over time. As a PhD student in 2013, Dr. Hammer published a paper that applied the Prototype-Willingness Model to study mental health help-seeking, in which he made the conceptual distinction between a person’s planned intention to seek help and their spontaneous willingness to seek help given conducive help-seeking circumstances. The idea is that someone might not possess an a priori intention to seek help, but if given the spontaneous convenient chance to do so, they might be willing to (e.g., a students sees a table in the campus student union advertising a free Let’s Talk drop in counseling opportunity with a university counseling center therapist and decides to try it out). However, having developed a stronger understanding of the reasoned action tradition and how it conceptualizes and measures intention, Dr. Hammer now sees less theoretical and practical value in the concept of “help-seeking willingness” as a separate construct worth operationalizing. Willingness is most likely to show empirical utility when intention is measured in a manner that makes the intention score obsolete with the passage of time, thereby allowing a willingness score to account for unique variance in prospective help-seeking behavior, because the intention score is outdated and thus will be a poorer predictor of prospective help-seeking behavior. In other words, willingness’s utility is mostly a parlor trick–a result of a research method limitation and statistical artifact. Here’s an example: Imagine we measure Sally’s intention to seek help on December 1st at 6am. Let’s pretend Sally scores low on intention at 6am, in part because she currently believes that she has no cost-effective source of mental health treatment available to her(a logistical belief that leads to low personal agency and thus low intention). But, let’s pretend that, later that day at 1pm, Sally learns that her employer covers 100% of the cost for a high-quality online therapy platform that Sally has heard great things about. In that moment, her beliefs about availability of accessible quality mental health treatment change, which results in higher personal agency, which results in greater intention to seek help. After work at 6pm, Sally logs on to this therapy platform and has a first session with a therapist (she engages in help-seeking behavior). If we measured Sally’s intention at 6am, her intention score would be a poor predictor of her prospective help-seeking behavior, and thus a help-seeking willingness item (e.g., “imagine you learned that your employer made high quality therapy available to you at no cost–to what degree would you be willing to seek mental health care?”) might demonstrate incremental predictive validity by being able to account for unique variance in prospective help-seeking behavior beyond what the 6am intention score could predict. But had we measured both intention and willingness at 1:01pm, when Sally’s true degree of intention increased due to a change in her help-seeking perceptions, her intention score would be a more accurate reflection of her true current degree of intention, and thus a stronger predictor of prospective help-seeking behavior, thereby making it much harder on the 1:01pm help-seeking willingness score to account for unique variance beyond the 1:01pm intention score. This example illustrates the importance of thinking carefully about the interdependent issues of measurement/research design, conceptual definitions, theoretical clarity, and practical applied utility. When researchers measure help-seeking intention in a manner that the score remains an accurate reflection of a person’s true latent degree of intention at a given point in time, the theoretical and empirical value of the willingness construct/measure is unclear. In summary, these days Dr. Hammer advises most people to focus on measuring intention in a valid manner and not trying to utilize willingness as an independent construct when studying mental health help-seeking. However, just because the construct of help-seeking willingness has limited value does not mean that researchers should avoid doing research on (1) spontaneous help-seeking decision making, (2) increasing spontaneous access to low-barrier mental health care services, (3) mental health help seeker prototypes, and (4) the role of “hot cognition” and emotion/affective factors in the help-seeking process. Lastly, it is not uncommon to see help-seeking researchers use the terms of “intention to seek help” and “willingness to seek help” interchangeably, and to see instruments that use one or both terms in the title/items of those measures; this interchangeable use is often harmless and fine. However, increasing care must be taken when we start to think about the fact that the way we conceptualize and operationalize intention in terms of the TACT-C elements (see Step 2 of this website’s Mixed-Method Protocol webpage for details) can start to influence our thinking about willingness vs intention. When we measure intention with conditional measures (e.g., If [hypothetical scenario, such as ‘I had a serious mental illness’], I would intent to seek help from a mental health professional in the next month”) versus unconditional measures (e.g., “I intend to seek help from a mental health professional in the next month”), this starts to raise questions about whether we’re measuring someone’s true current intention versus assessing their “willingness to seek help if X hypothetical scenario was true for them”. Note: this conceptual discussion is getting pretty complicated, so if does not make total sense, that’s totally understandable. Dr. Hammer is still working out the best way to think about these nuanced distinctions, and how best to communicate them to an external audience. It’s a work in progress 🙂