Sage Journals: Discover world-class research

Abstract

This article examines individuals’ likelihood of engaging in future health prediction as a function of their fatalism, future time orientation, superstition, and history of chronic disease. Using a multistage cluster sample of 33 urban cities in Turkey, we asked respondents (N = 1,467), to report their past and current health and predict their future (expected) health status (i.e., future self-rated health). While less than 1% failed to report past or current health, 23% of respondents provided no prediction for their future health status. We employed a moderated-mediation analysis to identify the predictors of this avoidance of reporting future health status expectations. Our analyses point to two potentially distinct mechanisms influencing individuals’ likelihood of providing future self-rated health. First, individuals suffering from a chronic disease were more likely to have higher fatalism, which, in turn, decreased their likelihood of providing a rating for their future health. Second, more superstitious individuals were less likely to report expectations about future health. This association was moderated by future time orientation such that for individuals with higher future time orientation (vs. present time orientation), higher superstition was associated with a steeper increase in the probability of avoidance of future health predictions. This finding suggests that some individuals might avoid sharing predictions about their future health because they fear talking about future outcomes can invite negative outcomes by “tempting fate.”

Plain language summary

This article looks at why some people might not want to report their predictions about their future health in surveys. We studied how people’s beliefs, thinking about the future, superstitions, and past health issues are connected to this tendency of some individuals to avoid reporting their expected future health. We asked people in 33 cities in Turkey about their health in the past, present, and what they expect in the future. Large majority of people answered the questions about their past and present health, but about a quarter didn’t answer the question about their future health. We found that people with ongoing health problems felt more fatalistic, and that made them less likely to predict their future health. Also, people who believed in superstitions were less likely to predict their future health, especially if they thought a lot about the future. This could be because they worried that talking about the future might bring bad luck.

Keywords

future self-rated health superstition tempting fate fatalism future time-orientation chronic illness

Imagine being asked the following in a questionnaire: “Looking ahead 5 years into the future, what do you expect your health will be like at that time?” (Ryff et al., 2015). Think about this item for a moment. What are the factors that may have a bearing on your evaluation? Possibly your current health comes to mind. Perhaps, you like to plan for the future and take measures for minimizing health-related risks. You might be the kind of person who believes that your fate is already written. You might even be the kind of person who lives in the moment and is not interested in what happens in the future. What if you made a prediction, but shied away from reporting it because you thought reporting your predictions would “temp fate” and bring bad luck?

Items such as the one above are used extensively to assess health status; they are simple in structure, easy to answer, and they successfully predict health outcomes such as mortality (e.g., Benyamini et al, 2000; Ferraro & Wilkinson, 2015). However, the thought processes such questions trigger may be complex and fraught with uncertainty for some individuals, resulting in non-response (Beatty et al., 1998). A more nuanced understanding of “don’t know” responses can be useful in terms of not only informing questionnaire design but also in terms of increasing the utility of self-rated health measures in the prediction of health status and identification of at-risk groups.

This study investigates factors associated with providing a “don’t know” response to such items to better identify who is unable or unwilling to report predictions about their future health status (FSRH; Future self-rated health). Specifically, we focus on individuals’ history of chronic disease, their health fatalism, superstition, and future time orientation as factors that may influence either the ability (cognitive state and adequacy judgments) or willingness (communicative intent) to respond to items measuring expectations about future health.

Self-Rated Health (SRH) as a Measure of Health Status

As a widely used measure, self-rated health (SRH) asks respondents to assess their health by responding to a single item, ranging from “worst” to “best” (e.g., Au & Johnston, 2014; Löckenhoff et al., 2012; Wang & Satariano, 2007). SRH has been employed in several major surveys, such as the World Values Survey (Inglehart et al., 2014) and the Midlife in the United States Survey (MIDUS) (Ryff et al., 2015).

A primary advantage of SRH is the ease with which it can be administered (Au & Johnston, 2014; Jylhä, 2009). It allows individuals to integrate a wide variety of information regarding their physical and mental health, well-being, and health risk factors (Huisman & Deeg, 2010; Sirois, 2015). Moreover, this assessment entails both an evaluation of one’s current health and expectations about a change in health status (Idler & Benyamini, 1997). As such, SRH can be useful in predicting future health behavior, such as the utilization of health services, and identifying risk factors, such as functional limitations and mortality (Deeg & Kriegsman, 2003; Reile et al., 2017). Given these considerations, a crucial area of inquiry concerns improving SRH by supplementing it with future self-rated health (FSRH, e.g., Sargent-Cox et al., 2010; Sirois, 2015). This line of research suggests that SRH and FSRH are only moderately correlated and they distinctly predict survival in the future (Sirois, 2015; Ferraro & Wilkinson, 2015). Furthermore, in line with the assumption that future health estimates are temporal projections that are partly based on current health, a combined measure of SRH and FSRH (e.g., good current health, bad future health) may be a stronger predictor of mortality than either SRH or FSRH (Wang & Satariano, 2007).

Predicting “Don’t Know” Responses to FSRH

A fundamental shortcoming of SRH is the lack of a clear understanding of what it measures (Jylhä, 2009; Layes et al., 2012). First, the holistic nature of SRH raises questions about what respondents were thinking (e.g., objective indicators vs. self-perceptions) when providing a specific rating (Au & Johnston, 2014; Layes et al., 2012). This problem becomes evident in light of findings suggesting that both individual differences, such as cognitive ability and optimism, and cross-cultural differences influence individuals’ assessment of their health (Black et al., 2017; Jylhä, 2009). Second, several factors, such as social norms and social acceptability of talking about health status, may influence the extent to which respondents will be willing or able to report their evaluations (Baron-Epel et al., 2005).

These two considerations are particularly relevant for a commonly encountered issue concerning SRH: use and treatment of “don’t know” responses. Many of the studies that utilize SRH either do not include “don’t know” as an option to respondents or do not explain whether “don’t know” option was provided and ran their analyses ignoring these responses (e.g., Ryff et al., 2015; Sirois, 2015). Lack of a “don’t know” option may result in a failure to distinguish between respondents who are unwilling versus who are unable to articulate their health status (Waters et al., 2013). Insofar as an inability to articulate health status is related to other health risk factors, such as health screening behavior (e.g., Orom et al., 2020; Waters et al., 2016), this may reduce the utility of SRH in the identification of risk groups and designing interventions. Furthermore, patterns of non-response to SRH items may be more pronounced among specific populations. For example, in the European Social Survey (ESS Round 7: European Social Survey Round 7 Data, 2014), the SRH item was incomplete for only 0.2% of all the respondents coming from 32 countries. However, a total of ten eastern European countries (e.g., Russia, Ukraine, Czech Republic, Turkey) accounted for 60% of the “don’t know” responses. Furthermore, “don’t know” responses may be particularly prevalent for individuals’ reporting of their future health status (FSRH). For example, in a study that compares the respective abilities of SRH and FSRH to predict mortality (Wang & Satariano, 2007), the rate of “don’t know” responses to the FSRH item was 17 times higher than that of the SRH items. As such, understanding the processes that explain “don’t know” responding to FSRH is crucial because of its implications for the identification of at-risk groups. For example, improving our understanding of these processes can help distinguish between non-response due to lack of knowledge, lack of motivation (both of which may be important indicators of risk), and other factors such as worldview or health cognitions (Waters et al. 2013; Waters et al., 2023).

“Can’t Tell” or “Don’t Want to Tell”?

According to the response basis framework, three main factors explain why individuals give “don’t know” responses in surveys (Beatty et al., 1998): cognitive state (i.e., whether respondents know the answer), adequacy judgment (i.e., respondents’ confidence about the answer), and communicative intent (i.e., whether respondents want to share the answer). For SRH, the cognitive state involves an evaluation of one’s objective health indicators as well as making judgments about one’s well-being (Huisman & Deeg, 2010; Jylhä, 2009). Concerning future self-rated health status, FSRH, the cognitive state may additionally entail consideration of one’s willingness or ability to make a projection about an uncertain future (Orom et al., 2020). Hence, it is likely that factors associated with the cognitive state for FSRH reporting will be different from that of SRH (Sirois, 2015). Below we discuss how individual differences such as fatalistic beliefs, superstition, future time-orientation may influence individuals’ ability and/or willingness to respond to questions asking about their future health status FSRH. Furthermore, we investigate whether fatalism and superstition are likely to be influenced by individuals’ current history of chronic health problems, defined as health problems that have “long duration and generally slow progression” (World Health Organization quoted in Bernell & Howard, 2016).

Fatalistic Beliefs

Given the uncertain nature of one’s future health prospects, it is likely that the extent to which respondents have a sense of control over their future will influence their tendency to provide a “don’t know” response to FSRH. This possibility is underlined by the Zimbardo time perspective inventory (Zimbardo & Boyd, 1999), which indicates that individuals with higher present-fatalistic orientation tend to believe that future outcomes are not influenced by individual action. This may have implications for both the cognitive state and adequacy judgments concerning FSRH. First, with respect to the former, fatalism, which refers to individuals’ beliefs that outcomes are predetermined by external forces and hence out of one’s control (Hui & Hui, 2009; Zhou et al., 2009), has been observed to reduce the extent to which an individual will engage in consideration of future outcomes (Zimbardo & Boyd, 1999).

At this point, it is essential to explain how the extent to which an individual will subscribe to fatalism will largely depend on prior experience—in this case, for example, whether an individual suffers from chronic problems. Namely, as predicted by the aforementioned dynamic evaluation thesis, individuals’ sense of control over future outcomes will be influenced by whether, in their current experiences, they have had the opportunity to do so (Franklin et al. 2007; Schafer et al., 2011). In line with these findings, research on survey response patterns suggest that while trying to save the day in the face of hardships, individuals will be more likely to focus on the present (e.g., present-fatalism) rather than think about future outcomes (Harper et al., 2003).

Second, with respect to adequacy judgments regarding one’s assessment of future health outcomes, individuals who are fatalistic may have a lower sense of future certainty (i.e., perceptions regarding the steadiness of one’s future) (Davis & Niebes-Davis, 2010), reducing their confidence in their ability to assess future health outcomes. While the consequences of this possibility have not been tested for future health predictions, it would imply that chronic health problems that continuously challenge individuals’ ability to control their health could decrease the likelihood that an individual is confident about their expectations about future health outcomes; this potentially reducing FSRH reporting.

Superstition

Fate control (perceived lack or ability to control fate) also comprises the belief that one can negotiate with fate (Chaturvedi et al., 2009; Hui & Hui, 2009; Kim et al., 2014). A quickly growing line of research suggests that one way through which individuals can reclaim agency over their fate is through superstition (e.g., Griffiths et al., 2019; Kim et al., 2014; Risen & Gilovich, 2008). Superstitious beliefs and behaviors (such as knocking on wood to stave bad luck) tend to assign the characteristics of one ontological category (i.e., a desired or an undesired outcome) to another category (e.g., thoughts, jinx, bad or good luck) (Berger, 2013; Lindeman & Svedholm, 2012; Risen, 2016; Vyse, 2013). An important finding that has consistently been observed in various domains, such as, health, and politics, is that individuals will be more likely to resort to superstitious behavior when they are under uncertain conditions (Hamerman & Morewedge, 2015; Stavrova & Meckel, 2017; Vyse, 2013; Zhang et al., 2014) to regain a sense of control and efficacy (Damisch et al., 2010; Risen, 2016).

To the extent that projecting future health involves an element of uncertainty, superstitious beliefs may be particularly germane to a respondent’s communicative intents related to FSRH. Namely, as proposed by the response basis framework, one important reason for negative communicative intent is lack of motivation. As we will outline next, superstitious beliefs, and specifically respondents’ tendency to avoid what has been named as “tempting fate” (e.g., Swirsky et al., 2011) may be a crucial factor that may reduce respondents’ motivation to report FSRH, even when they can make projections about their future.

“Tempting fate” refers to individuals’ perceptions that the probability of an outcome will be higher if they engage in an action that would tempt the outcome (Swirsky et al., 2011). A typical example of the tempting fate effect would be the belief that commenting on success would invite failure. This form of tempting fate effect refers to the belief that the act of being “arrogant” about one’s success, or more generally not acting in line with social norms related to accepting one’s limitations and being modest, would jinx the streak of success (Risen & Gilovich, 2008; Zhang et al., 2014). A closely related form of tempting fate occurs when individuals believe that mental counting of future outcomes may prevent achieving the outcome. Such an aversion against “counting one’s chickens before they hatch” reflects both a defensive pessimism (i.e., not being too enthusiastic so as to reduce disappointment) and a form of magical concern that thinking about the desired outcome will decrease the likelihood that it will happen (Pronin et al., 2006).

Several points are important to note about how tempting fate effect in the form of mental counting aversion may arise in the context of FSRH reporting. First, individuals may engage in such thoughts even if they think those thoughts are not logical (Pronin et al., 2006). The reason for this is what Risen (2016) names as the “acquiescence” phenomenon: individuals realize that their belief is not logical but stick with what their intuition suggests. FSRH reporting would be subject to such acquiescence because the perceived benefits from avoiding tempting fate by not reporting expected health would exceed the costs of not reporting FSRH. To our knowledge, there are no studies on how acquiescence may influence FSRH reporting (and propensity to provide “don’t know” response). However, according to Baron-Epel et al. (2005), differences in SRH reporting among Jewish and Arab populations in Israel may partly be due to differences in cultural beliefs regarding protecting oneself from evil-eye by presenting a negative picture of health (Baron-Epel et al., 2005).

Second, research indicates that even when inclined to think about the future outcome, tempting fate effect will mean that individuals will be reluctant to express their expectations (Krizan & Windschitl, 2009; Pronin et al., 2006; Risen & Gilovich, 2008). According to the response basis framework, such reluctance may result either in omission (whereby the respondent would provide a “don’t know” response) or an error of commission (whereby the respondents would provide false information about their expectations). In other words, the tempting fate effect underlines the possibility of a distinct process through which “don’t know” responding may occur. Namely, whereas for individuals with low future time orientation and high fatalism, “don’t know” responding may imply a lack of ability to do so, among individuals trying to avoid tempting fate, “don’t know” responding would imply a consideration of future outcomes.

Time Orientation

Time orientation, specifically future time orientation, is associated with careful consideration of the future including higher levels of future planning, heightened sensitivity to future consequences of their own current behavior, and avoidance of behaviors associated with future negative consequences (Zimbardo & Boyd, 1999). Individuals high in future time orientation actively engage in health-protective behaviors while minimizing risk-averse behaviors(Boyd & Zimbardo, 2005; Hall et al., 2014). As they are more likely to consider future outcomes than present-oriented individuals, it is not surprising that individuals with higher future time orientation have lower “don’t know” response rates for questions measuring expectations about future health (Orom et al., 2020). However, our discussion regarding superstition underlines the possibility that even among individuals with higher future orientation, communicative intent may be influenced by the extent to which individuals think talking about their expected future health would “tempt fate.” In other words, the association between time orientation and FSRH reported should be such that future time orientation should be more likely to lead to higher rates of “don’t know” response to FSRH among individuals with higher superstition.

Present Study

In this light, the current study aims to incorporate previously summarized empirical findings and examine factors associated with FSRH “don’t know” responding. We expect fatalism, superstition, chronic disease presence, and future time orientation to play key roles in “don’t know” responses. First, since fatalism and superstition are associated with lower perceptions of control over one’s health, they may also reduce individuals’ perceived ability to make accurate assessments about their future health, thereby increasing “don’t know” responses to FSRH. Second, having a history of chronic—recurring and difficult to treat—health problems would have an indirect negative effect on FSRH reporting by lowering future perceived sense of control over one’s health (e.g., increasing fatalism). Third, insofar as FSRH reporting entails both willingness and ability to consider future outcomes and communicative intent (willingness to report what one considers), we expect that future time orientation would have two potentially distinct effects on FSRH reporting.

On the one hand, we expect that those with higher future time orientation will, generally, be more likely to have the ability to consider future health outcomes and hence be less likely to provide a “don’t know” response to FSRH. On the other hand, even when able to consider future health outcomes, some individuals would refrain from communicating those expectations to avoid what has been referred to as “tempting fate.” Hence, we predict that there will be an interaction between future time orientation and superstition such that among those with higher future time orientation, the rate of “don’t know” responses will increase as superstition increases.

Methods

Sample

Data for this study were obtained from a larger study, conducted in Turkey, that aimed to investigate breast and cervical cancer screening among women and colorectal screening both among men and women. The broader study (N = 3,021) used multistage cluster sampling in 33 urban cities and involved face-to-face interviews lasting about an hour. The respondents were primarily female (80%) adults aged 20 and above (M = 44.19; SD = 13.63). As will be explained in the analytical approach section, the current study uses data from 1,467 individuals.

Measures

Health Ratings

Using three 11-point scale items (0 = worst possible health; 10 = best possible health) adopted from the MIDUS (Ryff et al. 2015), participants rated their current health (M = 6.99; SD = 1.83), health 5 years ago (M = 7.53; SD = 1.92), and expected health 5 years from now (M = 6.38; SD = 2.05). While “don’t know” responding was only 1% for SRH (current and past), it was 23% for FSRH.

Chronic Health Problems

Respondents noted whether they suffer from chronic health problem(s) that require follow-up and listed all such health problems. From these questions, the total number of chronic diseases that each respondent suffers from was computed (M = 0.4; SD = 0.73).

Fatalism

Participants answered three questions from the five-point Fatalism Scale (Shen et al., 2009): “If someone is meant to get a serious disease, they will no matter what they do,”“How long I live is predetermined,” and “I will die when I am fated to die” (α = .73; M = 3.94; SD = 0.85).

Superstitiousness

Using a four-point scale, respondents answered three questions regarding the extent to which they believed in: “being hexed,”“bad luck,” and “miracles” (α = .77; M = 3.13; SD = 0.73; Çarkoğlu & Çarkoğlu, 2009).

Future Time Orientation

Future time orientation was assessed using two reverse coded five-point scale items adopted from Strathman et al. (1994): “Must live in the moment, not think about future” and “No need to make long term plans” (r = .77, p < .001; M = 3.03; SD = 1.17).

Analytical Approach

We used a moderated-mediation analysis using the PROCESS (Hayes, 2017), a computational tool for conducting path-analysis-based tests of mediation, moderation and their combination (Hayes, 2017). Path-analysis provides a structured way to assess and quantify relationships between multiple variables, helping to identify direct and indirect effects, and providing insights into potential causal pathways among variables of interest. In this respect, path analysis allows us to assess how the individual characteristics described above may influence willingness to report predictions about future health. The PROCESS tool is also suitable for our purposes because it can estimate coefficients of a model using OLS regression for continuous outcomes and maximum likelihood logistic regression for dichotomous outcomes (such as the outcome variable in our study—whether or not a respondent engages in “don’t know” responding).

In this study, 23.8% of the respondents engaged in “don’t know” responding for FSRH, constituting what can be classified as a borderline rare outcome. For maximum likelihood logistic regression, the distribution of the outcome variable analyzed is key to their performance. Specifically, logistic regression tends to underestimate the probability of rare events (King & Zeng, 2003). In such cases, one commonly utilized solution to the problems discussed above is undersampling the cases representing the majority outcome (i.e., respondents providing a prediction FSRH) (Weiss, 2004). In the initial sample of the broader study (N = 3,021) 23.8% of the respondents did not provide an expectation about their future health (N = 719) and 76.2% did (N = 2,302). To create a balanced sample, we first split the sample into two groups using SPSS data selection procedure and then randomly selected 750 respondents from the group that provided an expectation about their future health. Two of the respondents from the randomly selected sub-group had missing data for the fatalism variable and hence were excluded. We then combined the two samples into a single dataset, resulting in a sample of 1,467 respondents in which 51% provided a prediction for their future health and 49% answering “don’t know.”

Results

Descriptive Summary

There was no significant difference between male and female respondents in terms of “don’t know” responding, χ² (1, N = 1,467) = 0.404, p = .53. Likewise, age was not significantly related to whether (M_age = 44, SD = 13.50) or not (M_age = 45, SD = 13.77) respondents provided “don’t know” as an answer for FSRH, F(1, 1,465) = .923, p = .34. The rate of “don’t know” responding was lower for respondents who had some high school education (46%) or were at least high school graduates (42.4%) than respondents who had primary school education (51.3%) or respondents who were not literate (60.7%), χ² (3, N = 1,467) = 15.477, p < .001.

Table 1 compares respondents who have provided a response to FSRH and respondents who responded “don’t know” with respect to self-rated health (5 years ago, present), chronic health problems, future time orientation, fatalism, and superstitiousness. When compared to respondents who provided a prediction for their future health, respondents who engaged in “don’t know” responding had lower ratings of health (for both past and current) and reported a higher number of chronic health problems. Also, participants providing “don’t know” as a response for FSRH had lower scores for future time orientation and higher scores for fatalism and superstitious beliefs.

Table 1.

Mean Differences.

	Prediction for future health provided		Prediction for future health not provided		p
Variable	M	SD	M	SD	p
Number of chronic health problems	0.33	0.61	0.56	0.85	.000
Self-rated health (now) (0–10)	7.00	1.85	6.73	1.85	.004
Self-rated health (5 years ago) (0–10)	7.56	1.89	7.32	2.07	.020
Future time orientation (1–5)	3.08	1.14	2.82	1.18	.000
Fatalism (predetermination) (1–5)	3.74	0.78	4.06	0.66	.000
Superstition (1–4)	3.07	0.75	3.25	0.74	.000

Moderated-Mediation Model

We conducted a moderated-mediation analysis with a bootstrap approach (5,000 drawings), using the PROCESS macro (Model 15) for SPSS (Hayes, 2017) with history of chronic problems as the independent variable, fatalism, and superstition as the mediating variables, future time orientation as a moderating variable and avoidance of future health prediction as the dependent variable (Figure 1).

Figure 1.

Moderated-mediation model predicting future health prediction avoidance.

The influence of history of chronic health problems on “don’t know” responding was mediated by fatalism (B = 0.08, p < .01, for the path from chronic health to fatalism; B = 0.71, odds ratio [OR] = 2.04, p < .001, for the path from fatalism to “don’t know” responding). After adding future time orientation as a moderating variable, the paths from history of chronic disease (B = 0.16, odds ratio [OR] = 1.175, p = .46) and superstitiousness (B = −0.27, odds ratio [OR] = 0.764, p = .18) to avoidance of future health prediction were no longer significant. However, there was a significant interaction between superstitiousness and future time orientation (CoxSnell = 0.87). Accordingly, as can be seen in Figure 2, the impact of an increase in superstition on the probability of “don’t know” reporting is stronger among respondents with higher future time orientation.

Figure 2.

Interaction between future time orientation and superstition.

Discussion

Individuals’ reporting of their health status (SRH) in surveys is commonly associated with socioeconomic factors that also influence their health, and relatedly, are predictive of health outcomes (e.g., McGee et al., 1999). Recent work has focused on improving SRH ratings by asking respondents also to predict their future health status (FSRH) (e.g., Sirois, 2015). The prevalence of non-response to FSRH items, particularly in the form of “don’t know” responses, presents a vital obstacle for their prognostic potential in terms of identifying risk groups. As such, this paper aimed to identify factors that predict “don’t know” responses to the FSRH item. Specifically, we focused on the history of chronic disease, fatalism, superstition, and future time orientation.

In line with our predictions, fatalism mediated the relationship between chronic disease and “don’t know” responding. Superstitiousness was not a mediating factor; however, it interacted with future time orientation in predicting “don’t know” response. For those high in superstitiousness, as future time orientation increased so did the likelihood of “don’t know” responding. That is, congruent with the notion of “tempting fate” we discussed in this paper, even when individuals (such as those who are high in future time orientation) are capable of projecting into the future, superstition—or more precisely the fear of tempting fate by voicing those projections—may reduce their communicative willingness, resulting in a higher rate of “don’t know” response.

Overall, in line with the response basis framework, these findings underscore the presence of two distinct mechanisms potentially influencing “don’t know” reporting. On the one hand, prior health problems, associated with lower perceptions of control over one’s health, may reduce willingness and ability to engage in projections about future health. On the other hand, the mechanism related to tempting fate is associated with higher future time orientation, suggesting that these respondents are providing a “don’t know” response despite (potentially) making projections about future health.

This distinction has important implications for the use of FSRH in identifying high-risk groups. The former type of “don’t know” responding, to the extent that perceived and actual lack of control is associated with a lower likelihood of adopting health-protective behaviors (e.g., Teo et al., 2016), may be an important predictor of future health problems. Conversely, the latter group, those with high superstition and high future time orientation, are more likely to be engaging in health-protective behavior. In this respect, due to their future planning and tendency to seek ways of reducing adverse consequences, this group may be less likely to be a high-risk group.

Additionally, insofar as such traits influence health risk, these findings are also indicative of how cross-cultural differences may influence the ways in which respondents react to the wording of questions. First, fatalism has been shown to be higher in immigrants and various minority ethnic groups such as Arabs in Israel and African-Americans and was also associated with their reduction in health preventive behaviors (Baron-Epel et al., 2005; Spurlock & Cullins, 2006). As many immigrant and minority groups are already marginalized and disadvantaged in terms of healthcare access and preventive medicine (Spallek et al., 2010), further exclusion from risk estimations and subsequent interventions would be even more detrimental. Second, Waters et al. (2013) observe that beliefs regarding tempting fate may be a factor that increases “don’t know” responding particularly when there is a mismatch between the risk frame provided by the question wording (western medical conceptualization of risk) and the understanding of risk within a given culture (when belief in the influence of higher powers on health is common). In this respect, further studies should investigate “culture-fair” (van de Vijver & Tanzer, 1997) approach to wording self-reported health-related questions and assessing risk.

It should be noted that the classification of individuals with high levels of “don’t know” responses within the framework of the response bias framework is somewhat speculative and would need further validation. Particularly, longitudinal studies would be needed to examine whether such classifications of “don’t know” responses predict differences in future health problems. Yet, the proposed differences in the mechanisms associated with “don’t know” responses to SRH (and particularly FSRH) items have important implications for design of questionnaires and how missing values in FSRH items should be treated. First, when FSRH items are used, it would be advisable to allow respondents to differentiate between not knowing and not wanting to report. Second, instead of conventional approaches like treating “don’t know” responses as missing data (which could potentially result in systematic exclusion of subgroups in a society) or pooling “don’t know” with “worse” responses (e.g., Wang & Satariano, 2007; this would potentially result in overestimation of health risks), an improved understanding of “don’t know” responses can help imputation of data used for predicting health risk. There are two key limitations of this study. First, the findings presented in this study come from a secondary analysis of a larger dataset within which we first observed the pattern about FSRH. Hence, we were only able to use relevant factors (e.g., superstition, fatalism, future time orientation) that were available in the dataset. For example, our measure of chronic health problems focused solely on the number of chronic health problems that each respondent reported and could not account for the effect of severity or intensity of such diseases on predictions of future health status. Second, while the path analysis we conducted is useful for identifying potential causal relationships and interactions, the data utilized for this study was cross-sectional and hence limits our ability to make inferences about causal mechanisms. Nevertheless, this study offers important insights that can help guide decisions about questionnaire design, particularly in terms of the need to distinguish between respondents who do not think about future health outcomes and respondents who think but avoid reporting their expectations about future health outcomes. This distinction is important because it underscores the existence of different health risk profiles. Specifically, insofar as not thinking about future outcomes is associated with a lower likelihood of taking measures to protect one’s health, the former type of respondents may benefit from interventions that invite them to think about their future health and take necessary actions. Given these insights, future work on improving the measurement of future health ratings is necessary both for methodological reasons and to help health practitioners in identifying at-risk groups.

Conclusion

This study offers insights into the interplay between individuals’ health reporting and socio-psychological factors. The investigation into self-rated health (SRH) and future self-rated health (FSRH) has revealed distinct mechanisms explaining “don’t know” responses. Fatalism mediates the connection between chronic disease history and such responses, while the interaction between superstitious tendencies and future time orientation unveils the impact of superstition on communication about health expectations. These findings have important implications for research methodology and healthcare strategies. Tailoring questionnaires to differentiate between those avoiding thoughts of future health and those who contemplate but refrain from reporting, along with recognizing cross-cultural influences, can enhance health-related assessments. While limitations exist, this study lays the groundwork for future research, inviting the exploration of causal mechanisms and the refinement of measurement tools to better understand health perceptions and promote effective interventions.

Footnotes

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study has been funded by The Scientific and Technological Research Council of Turkey (TÜBİTAK), Support Program 3501, Project No: 111K197.

Ethics Approval

The study was approved Koç University Committee on Evaluation of Ethical Issues (a copy of the approval and its English translation is available upon request).

Consent to Participate

Informed consent was obtained from all individual participants included in the study.

ORCID iDs

Lemi Baruh

Celia K. Naivar Sen

Zeynep Cemalcilar

G. Tarcan Kumkale

Data Availability

The datasets generated during and/or analysed during the current study are available in the OSF,

References

Johnston

D. W.

(2014). Self-assessed health: What does it mean and what does it hide? Social Science & Medicine, 121, 21–28.

Baron-Epel

Kaplan

Haviv-Messika

Tarabeia

Green

M. S.

Kaluski

D. N.

(2005). Self-reported health as a cultural health determinant in Arab and Jewish Israelis: MABAT - National Health and Nutrition Survey 1999–2001. Social Science & Medicine, 61, 1256–1266.

Beatty

Herrmann

Puskar

Kerwin

(1998). “Don’t know” responses in surveys: Is what I know what you want to know and do I want you to know it? Memory, 6, 407–426.

Benyamini

Idler

E. L.

Leventhal

E. A.

(2000). Positive affect and function as influences on self-assessments of health expanding our view beyond illness and disability. Journal of Gerontology: Psychological Sciences, 55(2), 107–116.

Berger

A. S.

(2013). The evil eye: A cautious look. Journal of Religion and Health, 52(3), 785–788.

Bernell

Howard

S. W.

(2016). Use your words carefully: What is a chronic disease? Frontiers in Public Health, 4, 159. https://doi.org/10.3389/fpubh.2016.00159

Black

Johnston

D. W.

Shields

M. A.

Suziedelyte

(2017). Who provides inconsistent reports of their health status? The importance of age, cognitive ability and socioeconomic status. Social Science & Medicine, 191, 9–18.

Boyd

J. N.

Zimbardo

P. G.

(2005). Time perspective, health, and risk taking. In Strathman

Joireman

(Eds.), Understanding behavior in the context of time: Theory, research, and application (pp. 85–107). Lawrence Erlbaum.

Çarkoğlu

(2009). Subjective well-being, perception of health and religiosity. Ankara.

10.

Chaturvedi

Chiu

C.-Y.

Viswanathan

(2009). Literacy, negotiable fate, and thinking style among low income women in India. Journal of Cross-Cultural Psychology, 40, 880–893.

11.

Damisch

Stoberock

Mussweiler

(2010). Keep your fingers crossed! How superstition improves performance. Psychological Science, 21, 1014–1020.

12.

Davis

M. J.

Niebes-Davis

A. J.

(2010). Ethnic differences and influence of perceived future certainty on adolescent and young adult sexual knowledge and attitudes. Health, Risk & Society, 12, 149–167.

13.

Deeg

D. J. H. H.

Kriegsman

D. M. W. W

. (2003). Concepts of self-rated health: Specifying the gender difference in mortality risk. The Gerontologist, 43, 376–386.

14.

ESS Round 7: European Social Survey Round 7 Data. (2014). Data file edition 2.2. NSD - Norwegian Centre for Research Data, Norway – Data Archive and Distributor of ESS Data for ESS ERIC. https://doi.org/10.21338/NSD-ESS7-2014.

15.

Franklin

M. D.

Schlundt

D. G.

McClellan

L. H.

Kinebrew

Sheats

Belue

Brown

Smikes

Patel

Hargreaves

(2007). Religious fatalism and its association with health behaviors and outcomes. Am J Health Behav, 31(6), 563–572. https://doi.org/10.5555/ajhb.2007.31.6.563

16.

Ferraro

K. F.

Wilkinson

L. R.

(2015). Alternative measures of self-rated health for predicting mortality among older people: Is past or future orientation more important? The Gerontologist, 55, 836–844.

17.

Griffiths

Shehabi

Murphy

R. A.

Le Pelley

M. E.

(2019). Superstition predicts perception of illusory control. British Journal of Psychology, 110, 499–518.

18.

Hall

P. A.

Fong

G. T.

Meng

(2014). Time perspective as a determinant of smoking cessation in four countries: Direct and mediated effects from the International Tobacco Control (ITC) 4-country surveys. Addictive Behaviors, 39, 1183–1190.

19.

Hamerman

E. J.

Morewedge

C. K.

(2015). Reliance on luck. Personality and Social Psychology Bulletin, 41, 323–335. https://doi.org/10.1177/0146167214565055

20.

Harber

Zimbardo

Boyd

(2003). Participant self-selection biases as a function of individual differences in time perspective. Basic and Applied Social Psychology, 25, 255–264. https://doi.org/10.1207/s15324834basp2503_08

21.

Hayes

(2017). Introduction to mediation, moderation, and conditional process analysis. The Guilford Press.

22.

Hui

C.-M.

Hui

H.-H. N.

(2009). The mileage from social axioms: Learning from the past and looking forward. In Kwok

Bond

M. H.

(Eds.), Psychological aspects of social axioms (pp. 13–30). Springer.

23.

Huisman

Deeg

D. J. H.

(2010). A commentary on Marja Jylhä’s “What is self-rated health and why does it predict mortality? Towards a unified conceptual model” (69:3, 2009, 307–316). Social Science & Medicine, 70, 652–654.

24.

Idler

E. L.

Benyamini

(1997). Self-rated health and mortality: A review of twenty-seven community studies. Journal of Health and Social Behavior, 38, 21–37.

25.

Inglehart

Haerpfer

Moreno

Welzel

Kizilova

Diez-Medrano

Lagos

Norris

Ponarin

Puranen

(2014). World values survey: Round six - country-pooled datafile version. JD Systems Institute. www.worldvaluessurvey.org/WVSDocumentationWV6.jsp

26.

Jylhä

(2009). What is self-rated health and why does it predict mortality? Towards a unified conceptual model. Social Science & Medicine, 69, 307–316.

27.

Kim

Kulow

Kramer

(2014). The interactive effect of beliefs in malleable fate and fateful predictions on choice. Journal of Consumer Research, 40, 1139–1148.

28.

King

Zeng

(2003). Logistic regression in rare events data. Journal of Statistical Software, 8, 137–163.

29.

Krizan

Windschitl

P. D.

(2009). Wishful thinking about the future: Does desire impact optimism? Social and Personality Psychology Compass, 3, 227–243.

30.

Layes

Asada

Kepart

(2012). Whiners and deniers - What does self-rated health measure? Social Science & Medicine, 75, 1–9.

31.

Lindeman

Svedholm

A. M.

(2012). What’s in a term? Paranormal, superstitious, magical and supernatural beliefs by any other name would mean the same. Review of General Psychology, 16, 241–255.

32.

Löckenhoff

C. E.

Terracciano

Ferrucci

Costa

P. T.

(2012). Five-factor personality traits and age trajectories of self-rated health: The role of question framing. Journal of Personality, 80, 375–401.

33.

McGee

D. L.

Liao

Cao

Cooper

R. S.

(1999). Self-reported health status and mortality in a multiethnic US cohort. American Journal of Epidemiology, 149, 41–46.

34.

Orom

Biddle

Waters

E. A.

Kiviniemi

M. T.

Sosnowski

A. N.

Hay

J. L.

(2020). Causes and consequences of uncertainty about illness risk perceptions. Journal of Health Psychology, 25(8), 1030–1042. https://doi.org/10.1177/1359105317745966

35.

Pronin

Wegner

D. M.

McCarthy

Rodriguez

(2006). Everyday magical powers: The role of apparent mental causation in the overestimation of personal influence. Journal of Personality and Social Psychology, 91, 218–231.

36.

Reile

Stickley

Leinsalu

(2017). Large variation in predictors of mortality by levels of self-rated health: Results from an 18-year follow-up study. Public Health, 145, 59–66.

37.

Risen

J. L.

(2016). Believing what we do not believe: Acquiescence to superstitious beliefs and other powerful intuitions. Psychological Review, 123, 182–207.

38.

Risen

J. L.

Gilovich

(2008). Why people are reluctant to tempt fate. Journal of Personality and Social Psychology, 95, 293–307. https://doi.org/10.1037/0022-3514.95.2.293

39.

Ryff

Almeida

Ayanian

Binkley

Carr

Coe

Davidson

Grzywacz

Karlamangla

Krueger

Lachman

(2015). National survey of midlife development in the United States (MIDUS 3), 2013–2014. Inter-university Consortium for Political and Social Research (ICPSR) [distributor].

40.

Sargent-Cox

K. A.

Anstey

K. J.

Luszcz

M. A.

(2010). Patterns of longitudinal change in older adults’ self-rated health: The effect of the point of reference. Health Psychology, 29, 143–152.

41.

Schafer

M. H.

Ferraro

K. F.

Mustillo

S. A.

(2011). Children of misfortune: Early adversity and cumulative inequality in perceived life trajectories. American Journal of Sociology, 116, 1053–1091.

42.

Shen

Condit

C. M.

Wright

(2009). The psychometric property and validation of a fatalism scale. Psychology and Health, 24, 597–613.

43.

Sirois

F. M.

(2015). Who looks forward to better health? Personality factors and future self-rated health in the context of chronic illness. International Journal of Behavioral Medicine, 22, 569–579. https://doi.org/10.1007/s12529-015-9460-8

44.

Spallek

Zeeb

Razum

(2010). Prevention among immigrants: The example of Germany. BMC Public Health, 10, 1–6.

45.

Spurlock

W. R.

Cullins

L. S.

(2006). Cancer fatalism and breast cancer screening in African American women. ABNF Journal, 17, 38–43.

46.

Stavrova

Meckel

(2017). The role of magical thinking in forecasting the future. British Journal of Psychology, 108, 148–168.

47.

Strathman

Gleicher

Boninger

D. S.

Edwards

C. S.

(1994). The consideration of future consequences: Weighing immediate and distant outcomes of behavior. Journal of Personality and Social Psychology, 66, 742–752.

48.

Swirsky

C. L.

Fernbach

P. M.

Sloman

S. A.

(2011). An illusion of control modulates the reluctance to tempt fate. Judgment & Decision Making, 6, 688–696.

49.

Teo

C. H.

C. J.

Booth

White

(2016). Barriers and facilitators to health screening in men: A systematic review. Social Science & Medicine, 165, 168–176. https://doi.org/10.1016/j.socscimed.2016.07.023

50.

van de Vijver

F. J. R.

Tanzer

N. K.

(1997). Bias and equivalence in cross-cultural assessment: An overview. European Review of Applied Psychology, 47, 263–280.

51.

Vyse

S. A.

(2013). Believing in magic: The psychology of superstition. Oxford University Press.

52.

Wang

Satariano

W. A.

(2007). Self-rated current and future health independently predict subsequent mortality in an aging population. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 62, 1428–1434.

53.

Waters

E. A.

Hay

J. L.

Orom

Kiviniemi

M. T.

Drake

B. F.

(2013). “Don’t Know” responses to risk perception measures: Implications for underserved populations. Medical Decision Making, 33, 271–281.

54.

Waters

E. A.

Kiviniemi

M. T.

Orom

Hay

J. L.

(2016). “I don’t know” my cancer risk: Implications for health behavior engagement. Annals of Behavioral Medicine, 50, 784–788.

55.

Waters

E. A.

Pogge

Orom

Kiviniemi

M. T.

Hay

J. L.

Lewicka

Allard

N. C.

Webster

G. D.

Shepperd

J. A.

(2023). I don’t know my child’s asthma risk: evidence against satisficing as an explanation for ‘don’t know’ responses. Journal of Risk Research, 26(12), 1370–1382. https://doi.org/10.1080/13669877.2023.2288006

56.

Weiss

(2004). Mining with rarity: A unifying framework. SIGKDD Explorations, 6(1), 7–19. https://doi.org/10.1145/1007730.1007734

57.

Zhang

Risen

J. L.

Hosey

(2014). Reversing one’s fortune by pushing away bad luck. Journal of Experimental Psychology: General, 143, 1171–1184.

58.

Zhou

Leung

Bond

M. H.

(2009). Social axioms and achievement across cultures: The influence of reward for application and fate control. Learning and Individual Differences, 19, 366–371.

59.

Zimbardo

P. G.

Boyd

J. N.

(1999). Putting time in perspective: A valid, reliable individual-differences metric. Journal of Personality and Social Psychology, 77, 1271–1288.

Let’s Not Tempt Fate: The Influence of Future Time-Orientation,Fatalism,and Superstition on Willingness to Report Expectations about Future Health

Abstract

Plain language summary

Keywords

Self-Rated Health (SRH) as a Measure of Health Status

Predicting “Don’t Know” Responses to FSRH

“Can’t Tell” or “Don’t Want to Tell”?

Fatalistic Beliefs

Superstition

Time Orientation

Present Study

Methods

Sample

Measures

Health Ratings

Chronic Health Problems

Fatalism

Superstitiousness

Future Time Orientation

Analytical Approach

Results

Descriptive Summary

Moderated-Mediation Model

Discussion

Conclusion

Footnotes

Declaration of Conflicting Interests

Funding

Ethics Approval

Consent to Participate

ORCID iDs

Data Availability

References