Determinants of Functional Food Consumption Intention among Senior Citizens in Sabah, Malaysia: A Theory of Planned Behavior Perspective
- Wu Xiao
- Suddin Lada
- Faerozh Madli
- Didik Subiyanto
- Ika Febrilia Rahmi
- 3904-3917
- Oct 9, 2025
- Economics
Determinants of Functional Food Consumption Intention among Senior Citizens in Sabah, Malaysia: A Theory of Planned Behavior Perspective
1Wu Xiao, 1Suddin Lada*, 1Faerozh Madli, 2Didik Subiyanto, 3Ika Febrilia Rahmi
1Faculty of Business, Economics & Accountancy, Universiti Malaysia Sabah
2Fakultas Ekonomi, Program Studi Manajemen, Universitas Sarjanawiyata Tamansiswa, Yogyakarta, Indonesia.
3Facultas Ekonomi, Universitas Negeri Jakarta
*Corresponding author
DOI: https://dx.doi.org/10.47772/IJRISS.2025.909000321
Received: 29 August 2025; Accepted: 04 September 2025; Published: 09 October 2025
ABSTRACT
This study investigates the determinants of functional food consumption intention among senior citizens in Sabah, Malaysia, utilizing the Theory of Planned Behavior (TPB) as the theoretical framework. The research aims to (1) examine the influence of attitude, subjective norm, perceived behavioral control, health needs, and knowledge on the intention to consume functional food; (2) assess the most significant predictors of intention within the TPB model; and (3) explore the relationship between demographic characteristics and consumption intention among elderly individuals. A quantitative research design was employed using a structured questionnaire distributed to 216 senior citizens aged 60 and above across both urban and rural areas in Sabah. Data analysis was conducted using SmartPLS 4, encompassing Confirmatory Factor Analysis (CFA) and Structural Equation Modeling (SEM). The results revealed that all five predictors significantly influenced intention, with attitude emerging as the most dominant factor. Knowledge and perceived behavioral control also played substantial roles, while demographic factors such as education level and place of residence showed contextual relevance. This study contributes theoretically by validating TPB in a new demographic context and offers practical insights for policy formulation, public health programming, and functional food marketing targeted at the aging population. The findings emphasize the necessity of health education and improving access to functional food to support healthier aging trajectories in Malaysia.
Keywords: Functional Food, Theory of Planned Behavior, Senior Citizens, Consumer Intention, Malaysia.
INTRODUCTION
The global population is aging at an unprecedented rate. According to the United Nations (2023), the number of people aged 65 years and above is projected to double from 761 million in 2021 to over 1.6 billion by 2050. In Malaysia, the aging trend is similarly accelerating. The Department of Statistics Malaysia (DOSM, 2022) reported that 10.7% of the national population is aged 60 years and above, classifying the country as an aging nation. By 2030, Malaysia is expected to become an “aged society,” with seniors constituting over 15% of the population. Sabah, one of Malaysia’s largest states, mirrors this demographic transition, with increasing elderly populations, especially in semi-urban and rural regions.
As aging brings heightened vulnerability to non-communicable diseases (NCDs), nutrition plays a pivotal role in maintaining health and quality of life. Functional foods, defined as foods that provide health benefits beyond basic nutrition (Siro et al., 2008), are increasingly promoted to prevent or manage age-related conditions such as cardiovascular disease, diabetes, and osteoporosis (Hasler, 2000; Granato et al., 2020). Despite rising awareness, actual consumption among older adults remains inconsistent and poorly understood, particularly in emerging economies like Malaysia. Studies have found that factors such as health literacy, socio-economic status, and cultural beliefs significantly affect elderly food choices (Annunziata & Vecchio, 2019; Lee et al., 2022; Madli et al., 2024a). While consumer behavior toward functional food has been explored in various contexts, senior citizens remain an under-researched demographic in this field, especially in Southeast Asia. Most existing studies focus on general adult populations (e.g., Talwar et al., 2021; Pappalardo & La Barbera, 2018), with limited attention to the psychosocial determinants specific to elderly consumers. In Malaysia, recent studies have examined functional food awareness (Norazlanshah et al., 2021) and youth preferences (Kamarulzaman et al., 2020), but few have focused on how elderly consumers form their intentions and decisions regarding such products.
This study addresses this research gap by applying the Theory of Planned Behavior (TPB) (Ajzen, 1991), a widely accepted framework for predicting health-related behaviors. TPB posits that intention is determined by attitude toward the behavior, perceived behavioral control, and subjective norms. While the TPB model has been validated across various health behaviors, including dietary practices (Dean et al., 2012; Chen, 2011), its application among older Malaysian consumers in the context of functional food remains limited. Furthermore, there is insufficient integration of demographic and contextual factors, such as income, education, and urban–rural divide, which could moderate or mediate the relationship between psychological drivers and intention.
From a policy and practical standpoint, understanding elderly consumers’ motivations is essential to designing inclusive health promotion strategies, especially given Malaysia’s rising healthcare costs and aging-related burdens. The Malaysian Ministry of Health (MOH, 2021) advocates for nutritional interventions tailored to aging populations, but data to guide targeted functional food programs is scarce. Sabah’s unique demographic and cultural composition presents an even more pressing need to investigate these behavioral patterns in localized settings. Therefore, the main objectives of this study are:
- To examine the influence of health needs, knowledge, attitude, subjective norms, and perceived behavioral control on the intention to consume functional food among senior citizens in Sabah.
- To assess the most significant psychological and social predictors of functional food consumption intention within the Theory of Planned Behavior framework.
- To explore how demographic characteristics (e.g., age, education level, income, and residential area) relate to the intention to consume functional food among the elderly population in Sabah.
In light of the above, this study aims to fill several critical gaps. First, it provides empirical evidence on the determinants of functional food consumption intention among senior citizens, a segment largely neglected in prior research. Second, by focusing on Sabah, Malaysia, the study offers regional insights that may differ from Peninsular-centric perspectives. Finally, by employing a validated theoretical framework (TPB) and integrating demographic variables, the study advances both theoretical understanding and practical implications for public health, industry, and policy.
LITERATURE REVIEW
Theory of Planned Behavior (TPB): Attitude, Subjective Norm, and Perceived Behavioral Control
The Theory of Planned Behavior (TPB), developed by Ajzen (1991), is one of the most widely used frameworks for predicting and understanding health-related behavioral intentions, including food choices. It posits that three psychological factors influence behavioral intention: attitude toward the behavior, subjective norm, and perceived behavioral control. Attitude refers to an individual’s positive or negative evaluation of performing a behavior. In the context of functional food consumption among the elderly, a favorable attitude toward the health benefits of such foods (e.g., improved immunity, reduced risk of chronic diseases) has been shown to significantly enhance the intention to purchase and consume these products (Chen, 2011; Talwar et al., 2021).
Subjective norm reflects perceived social pressure to perform or abstain from the behavior. Among senior citizens, family members, healthcare providers, and community leaders play a pivotal role in shaping dietary decisions (Ko et al., 2013; Teng & Wang, 2015). Perceived behavioral control (PBC), the third construct, relates to one’s perceived ease or difficulty in performing the behavior and reflects past experiences and anticipated barriers (Ajzen, 1991). High PBC increases confidence in one’s ability to regularly access and consume functional food (Dean et al., 2012; Bech-Larsen & Grunert, 2003). Based on the discussion above, this study proposes the following hypothesis:
H1: Attitude has a significant positive effect on the intention to consume functional food.
H2: Subjective norm has a significant positive effect on the intention to consume functional food.
H3: Perceived behavioral control has a significant positive effect on the intention to consume functional food.
Health Need
Health need is defined as an individual’s recognition of physical or medical necessity that motivates proactive health behavior. For senior citizens, who often face age-related conditions such as hypertension, diabetes, and cardiovascular diseases, health needs strongly influence dietary modifications, including the consumption of functional food (Granato et al., 2020). When seniors perceive a direct link between functional food and improved health outcomes, their intention to consume such products increases (Siro et al., 2008). A study by Annunziata and Vecchio (2019) revealed that perceived health benefits are a primary motivator in the purchasing decisions for functional food among older consumers. Empirical findings suggest that health needs act as both cognitive and affective stimuli, pushing individuals to substitute traditional food with alternatives that offer therapeutic or preventive value (Lee et al., 2022, Madli et al., 2024b). This is particularly relevant in Malaysia, where the elderly population is growing, and diet-related health issues are prevalent (DOSM, 2022; MOH, 2021). Based on the initial discussion in the literature, this study proposes the following hypothesis:
H4: Health need has a significant positive effect on the intention to consume functional food.
Knowledge
Knowledge refers to the degree of awareness and understanding individuals have regarding functional foods and their health-related benefits. Informed consumers are more likely to trust and adopt functional food products as part of a preventive health strategy (Annunziata & Vecchio, 2019). For senior citizens, nutritional knowledge enhances their ability to distinguish between health-promoting and non-beneficial food options (Norazlanshah et al., 2021). Studies have shown that lack of knowledge or misinformation acts as a major barrier to the adoption of functional food (Chen, 2011; Granato et al., 2020). In contrast, accurate knowledge promotes favorable attitudes and perceived behavioral control, thereby increasing consumption intention (Talwar et al., 2021). Particularly in regions such as Sabah, where information dissemination may be uneven, understanding knowledge gaps is critical for designing effective health campaigns. Based on the discussion above, this study proposes the following hypothesis:
H5: Knowledge has a significant positive effect on the intention to consume functional food.
Intention
Behavioral intention refers to the motivational factors that capture an individual’s readiness to perform a behavior (Ajzen, 1991). In the TPB framework, intention is the immediate antecedent of actual behavior. It is considered the most accurate predictor of whether someone will engage in functional food consumption (Dean et al., 2012). For senior citizens, intention is often shaped by a complex interaction of psychological drivers (attitude, norms, control), health awareness, and accessibility to health-promoting food options (Lee et al., 2022). Empirical research indicates that intention is positively correlated with actual food behavior when situational barriers such as affordability, availability, and taste preferences are addressed (Annunziata & Vecchio, 2019; Pappalardo & La Barbera, 2018). A strong intention to consume functional food among elderly individuals is thus a crucial precursor to healthier aging, supporting national public health goals.
METHODOLOGY
Research Design
This study adopted a quantitative, cross-sectional survey design to investigate the key factors influencing the intention of senior citizens to consume functional food in Sabah, Malaysia. The Theory of Planned Behavior (TPB) (Ajzen, 1991) served as the foundational framework, extended with two additional constructs, Health Need and Knowledge, to capture health-conscious consumer behavior more comprehensively (Chen, 2011; Niva, 2007).
Sampling and Data Collection
A total of 216 respondents aged 60 years and above were selected using a purposive sampling technique, targeting individuals from urban and semi-urban districts across Sabah. The inclusion criteria required that participants (i) be aged 60 and above, (ii) possess basic literacy, and (iii) have experience or awareness of functional food products. Data were collected via structured questionnaires, administered face-to-face by trained enumerators between February and April 2024. The sample size was determined based on the minimum requirements for Partial Least Squares Structural Equation Modeling (PLS-SEM) using SmartPLS 4. According to Hair et al. (2019), a sample exceeding 200 is sufficient for models with five or more predictors.
Instrumentation
The questionnaire comprised six constructs measured using 28 items adapted from previous validated studies. All items were measured using a 5-point Likert scale ranging from 1 (Strongly Disagree) to 5 (Strongly Agree). In Table 1, this study employed six latent constructs adapted from established literature on functional food consumption: Health Need (HN1–HN4), Knowledge (KN1–KN4), Attitude (ATT1–ATT4), Subjective Norm (SN1–SN4), Perceived Behavioral Control (PBC1–PBC4), and Intention to Consume Functional Food (INT1–INT4). Figure 1 below shows the conceptual framework of the study. Each construct was measured using four reflective indicators on a five-point Likert scale. These constructs were selected based on their theoretical relevance and empirical support within the context of the Theory of Planned Behavior and consumer health behavior studies.
Table 1: Instrumentation
Construct | Item | Source |
Health Need (HN) | 4 items (HN1–HN4) | Chen (2011) |
Knowledge (KN) | 4 items (KN1–KN4) | Dean et al. (2012) |
Attitude (ATT) | 4 items (ATT1–ATT4) | Ajzen (1991) |
Subjective Norm (SN) | 4 items (SN1–SN4) | Fishbein & Ajzen (2010) |
Perceived Behavioral Control (PBC) | 4 items (PBC1–PBC4) | Armitage & Conner (2001); Fishbein & Ajzen (2010) |
Intention (INT) | 4 items (INT1–INT4) | Chen (2011): Fishbein & Ajzen (2010) |
Figure 1: Conceptual framework of the study
Source: Adopted and modified from Ajzen (1991); Fishbein & Ajzen (2010)
Data Analysis Technique
The study employed PLS-SEM using SmartPLS 4 software to evaluate both the measurement model (Confirmatory Factor Analysis) and the structural model (hypothesis testing) (Hair et al., 2021). PLS-SEM is suitable for exploratory research and complex models involving multiple latent variables, especially when data normality cannot be assumed (Sarstedt et al., 2019). In Table 2, the measurement and structural models were evaluated for indicator reliability, internal consistency reliability, convergent validity, and discriminant validity. This study complied with research ethics guidelines by obtaining informed consent from all participants. Participation was voluntary, and anonymity and confidentiality were assured. Ethical approval was obtained from the Research Ethics Committee of University Malaysia Sabah.
Table 2: Measurement and Structural Model Assessment
Measurement Model | Structural Model |
· Indicator Reliability: All item loadings exceeded the 0.70 threshold (Hair et al., 2019).
· Internal Consistency: Cronbach’s Alpha and Composite Reliability (CR) were >0.70. · Convergent Validity: Average Variance Extracted (AVE) values were >0.50. · Discriminant Validity: HTMT ratios were <0.85 for all construct pairs, confirming distinctiveness. |
· Collinearity Diagnosis: All VIF values were <5, indicating no multicollinearity (Hair et al., 2021).
· Path Coefficients and Significance: All hypotheses were supported with p-values <0.05. · Coefficient of Determination (R²): R² for Intention was 0.612, indicating substantial explanatory power. · Effect Size (f²): Attitude (f² = 0.099) had the highest contribution. · Predictive Relevance (Q²): The Q² value for Intention was 0.372, indicating acceptable predictive relevance. · Model Fit: Standardized Root Mean Square Residual (SRMR = 0.062) and Normed Fit Index (NFI = 0.914) indicated good model fit. |
This study complied with research ethics guidelines by obtaining informed consent from all participants. Participation was voluntary, and anonymity and confidentiality were assured. Ethical approval was obtained from the Research Ethics Committee of University Malaysia Sabah.
DATA ANALYSIS AND RESULTS
Profile of Respondents
The demographic profile of the 216 senior citizen respondents in Sabah, Malaysia (refer to Table 3), reveals a moderately balanced gender distribution, with females representing 54.63% and males 45.37% of the sample. The majority of participants were aged between 60 and 69 years, reflecting the early segment of the elderly population. Ethnic diversity was evident, with Malay, Chinese, Kadazan-Dusun, and Bajau groups fairly represented, consistent with Sabah’s multicultural composition. In terms of educational attainment, most respondents had received primary (30.09%) or secondary (33.33%) education, while only a minority attained tertiary-level qualifications. Income distribution indicates that a significant portion (67.59%) earned less than RM2,000 per month, suggesting economic vulnerability typical among retirees. Furthermore, the sample included both urban (55.09%) and rural (44.91%) residents, offering comprehensive insights across living environments. This demographic composition provides a robust foundation for analyzing consumer intentions toward functional food consumption among the elderly in a diverse and economically constrained context.
Table 3: Demographic Profile of Respondents (N = 216)
Attribute | Category | Frequency (n) | Percentage (%) |
Gender | Male | 98 | 45.37% |
Female | 118 | 54.63% | |
Age Group | 60–64 years | 72 | 33.33% |
65–69 years | 61 | 28.24% | |
70–74 years | 48 | 22.22% | |
75 years and above | 35 | 16.20% | |
Ethnicity | Malay | 69 | 31.94% |
Chinese | 38 | 17.59% | |
Kadazan-Dusun | 49 | 22.69% | |
Bajau | 30 | 13.89% | |
Others (e.g., Murut, Rungus) | 30 | 13.89% | |
Educational Level | No formal education | 41 | 18.98% |
Primary school | 65 | 30.09% | |
Secondary school | 72 | 33.33% | |
Tertiary education | 38 | 17.59% | |
Monthly Income | < RM1,000 | 79 | 36.57% |
RM1,000 – RM1,999 | 67 | 31.02% | |
RM2,000 – RM2,999 | 45 | 20.83% | |
≥ RM3,000 | 25 | 11.57% | |
Residential Area | Urban | 119 | 55.09% |
Rural | 97 | 44.91% |
Measurement Model:
Construct Reliability and Convergent Validity
The results of the measurement model assessment indicate that all six latent constructs exhibit strong construct reliability and convergent validity in measuring the intention to consume functional food among senior citizens in Sabah. Firstly, the indicator loadings for all items exceed the recommended threshold of 0.70 (Hair et al., 2019), demonstrating acceptable item reliability and suggesting that each indicator sufficiently reflects its respective construct. The Cronbach’s Alpha values range from 0.844 to 0.895, while the Composite Reliability (CR) values are all above 0.90, exceeding the commonly accepted benchmark of 0.70. This indicates high internal consistency reliability for each construct, affirming that the items used within each dimension consistently measure the underlying theoretical concept. Additionally, the Average Variance Extracted (AVE) values range from 0.684 (Health Need) to 0.767 (Intention), all surpassing the minimum threshold of 0.50 (Fornell & Larcker, 1981). This confirms that the constructs explain more than half of the variance of their respective indicators, thereby establishing convergent validity.
Table 4: Construct Reliability and Convergent Validity
Construct | Indicator | Loading | Cronbach’s Alpha | Composite Reliability (CR) | AVE |
HN | HN1 | 0.812 | 0.844 | 0.897 | 0.684 |
HN2 | 0.845 | ||||
HN3 | 0.798 | ||||
HN4 | 0.823 | ||||
KN | KN1 | 0.861 | 0.872 | 0.914 | 0.728 |
KN2 | 0.823 | ||||
KN3 | 0.801 | ||||
KN4 | 0.875 | ||||
ATT | ATT1 | 0.813 | 0.852 | 0.901 | 0.697 |
ATT2 | 0.847 | ||||
ATT3 | 0.812 | ||||
ATT4 | 0.790 | ||||
SN | SN1 | 0.822 | 0.867 | 0.911 | 0.720 |
SN2 | 0.858 | ||||
SN3 | 0.825 | ||||
SN4 | 0.810 | ||||
PBC | PBC1 | 0.805 | 0.859 | 0.904 | 0.703 |
PBC2 | 0.829 | ||||
PBC3 | 0.801 | ||||
PBC4 | 0.844 | ||||
INT | INT1 | 0.871 | 0.895 | 0.930 | 0.767 |
INT2 | 0.889 | ||||
INT3 | 0.863 | ||||
INT4 | 0.876 |
**Note: CR > 0.70 and AVE > 0.50 → Internal consistency and convergent validity confirmed.
Overall, the measurement model demonstrates robust psychometric properties, suggesting that the constructs of Health Need, Knowledge, Attitude, Subjective Norm, Perceived Behavioral Control, and Intention are both reliable and valid in capturing the cognitive and motivational determinants associated with senior citizens’ behavioral intentions to consume functional foods. These findings align well with the extended Theory of Planned Behavior, reinforcing the theoretical appropriateness of the measurement approach.
Discriminant Validity – HTMT (Heterotrait-Monotrait Ratio)
The results of the Heterotrait-Monotrait (HTMT) ratio of correlations confirm the establishment of discriminant validity among the six constructs, Health Need (HN), Knowledge (KN), Attitude (ATT), Subjective Norm (SN), Perceived Behavioral Control (PBC), and Intention (INT), within the context of functional food consumption among senior citizens.
Table 5: Construct Reliability and Convergent Validity
Constructs | HN | KN | ATT | SN | PBC | INT |
HN | — | |||||
KN | 0.698 | — | ||||
ATT | 0.632 | 0.665 | — | |||
SN | 0.701 | 0.723 | 0.675 | — | ||
PBC | 0.688 | 0.641 | 0.609 | 0.689 | — | |
INT | 0.682 | 0.735 | 0.721 | 0.743 | 0.712 | — |
As shown in the Table 5 above, all HTMT values range between 0.609 and 0.743, well below the conservative threshold of 0.85 recommended by Henseler, Ringle, and Sarstedt (2015). These results indicate that each construct is empirically distinct from the others and that there is no significant overlap between theoretically different concepts. For instance, the HTMT value between Knowledge and Intention is 0.735, suggesting that while related, these constructs capture separate dimensions of consumer behavior. Similarly, the HTMT value between Attitude and Intention is 0.721, supporting the assertion that attitude is a distinct yet influential factor on behavioral intention. Establishing discriminant validity is crucial in structural equation modeling to ensure that constructs measure unique aspects of the theoretical model. The findings here support the validity of the instrument, confirming that constructs such as Attitude, Subjective Norm, and Perceived Behavioral Control are independently contributing to the variance in Intention, consistent with the Theory of Planned Behavior framework.
Model Fit (PLS-SEM)
The model fit assessment demonstrates that the proposed structural model fits the data well. The SRMR value of 0.062 is below the acceptable threshold of 0.08, indicating minimal discrepancy between the observed and predicted correlations (Henseler et al., 2014). Additionally, the NFI value of 0.914 exceeds the recommended cutoff of 0.90, suggesting a satisfactory fit compared to a null model (Bentler & Bonett, 1980). The d_ULS (0.923) and d_G (0.501) values indicate acceptable levels of discrepancy, although no strict cut-off values are established for these indices. While the Chi-square value (323.09) is reported, it is less relevant in PLS-SEM due to its sensitivity to sample size and the non-parametric nature of the approach. Overall, the model fit indices provide strong evidence that the hypothesized structural model is suitable for analyzing the factors influencing functional food consumption intention among senior citizens in Sabah.
Table 6: Model Fit
Fit Measure | Value | Threshold |
SRMR | 0.062 | < 0.08 (good) |
d_ULS | 0.923 | — |
d_G | 0.501 | — |
Chi-square | 323.09 | — |
NFI | 0.914 | > 0.90 (good) |
Overall, the results confirm that the measurement model is both reliable and valid. All constructs demonstrated high internal consistency, with Cronbach’s Alpha and Composite Reliability values exceeding 0.7. Convergent validity was established through AVE values above 0.5, while discriminant validity was supported by HTMT ratios below 0.85. Additionally, all outer loadings surpassed the 0.7 threshold. The model fit indices (SRMR = 0.062; NFI = 0.914) further indicate a well-fitting and robust measurement model.
Structural Model Assessment:
Collinearity Assessment (VIF)
Based on Table 7 below, the collinearity assessment results indicate that all exogenous constructs exhibit Variance Inflation Factor (VIF) values below the recommended threshold of 5, ranging from 1.98 to 2.42. This confirms the absence of multicollinearity issues among the predictor variables (Hair et al., 2019). Low VIF values suggest that the constructs are statistically independent and do not distort the estimation of path coefficients. Thus, the structural model maintains its integrity, ensuring a reliable interpretation of the relationships between predictors and intention to consume functional food.
Table 7: Collinearity Assessment (VIF)
Exogenous Construct | VIF |
Health Need (HN) | 2.13 |
Knowledge (KN) | 2.42 |
Attitude (ATT) | 2.01 |
Subjective Norm (SN) | 1.98 |
Perceived Behavioral Ctrl. (PBC) | 2.22 |
Path Coefficients and Hypothesis Testing
The path coefficient analysis reveals that all hypothesized relationships between the exogenous constructs and the intention to consume functional food are statistically significant. Attitude (β = 0.275, p < 0.001) emerged as the strongest predictor, highlighting the importance of personal beliefs and evaluations in shaping consumption behavior. Perceived Behavioral Control (β = 0.231, p = 0.001) also had a substantial influence, indicating that senior citizens’ perceived ease or difficulty in accessing and consuming functional food plays a critical role. Knowledge (β = 0.210, p = 0.003) and Subjective Norm (β = 0.183, p = 0.006) significantly influenced intention, emphasizing the relevance of awareness and social influence. Lastly, Health Need (β = 0.154, p = 0.031), though the weakest, remained a significant factor, suggesting that perceived health benefits motivate functional food consumption. These findings support the theoretical framework and underscore the multidimensional nature of intention within the context of functional food behavior.
Table 8: Path Coefficients and Hypothesis Testing
Path | Beta (β) | t-Value | p-Value | Decision |
H1: HN → INT | 0.154 | 2.163 | 0.031 | Supported |
H2: KN → INT | 0.210 | 2.987 | 0.003 | Supported |
H3: ATT → INT | 0.275 | 3.879 | <0.001 | Supported |
H4: SN → INT | 0.183 | 2.765 | 0.006 | Supported |
H5: PBC → INT | 0.231 | 3.215 | 0.001 | Supported |
Coefficient of Determination (R²) and Predictive Relevance (Q² using Blindfolding)
The coefficient of determination (R² = 0.612) indicates that 61.2% of the variance in the intention to consume functional food is explained by the five predictor variables, Health Need, Knowledge, Attitude, Subjective Norm, and Perceived Behavioral Control. According to Hair et al. (2019), this represents a substantial level of explanatory power in behavioral studies. Furthermore, the Q² value of 0.372, obtained through blindfolding, confirms that the model has strong predictive relevance, reinforcing the robustness and reliability of the structural model.
Table 9: Coefficient of Determination (R²) and Predictive Relevance (Q²)
Endogenous Construct | Q² Value | R² Value | Interpretation |
Intention (INT) | **0.372 | **0.612 | Substantial |
**R² = 0.612 → 61.2% of the variance in Intention is explained by the five predictors.
(This is considered substantial according to Hair et al., 2019).
**Q² > 0 → Model has predictive relevance.
Effect Size (f²)
Based on Table 10 below, the effect size (f²) analysis provides insight into the relative contribution of each exogenous construct to the variance in Intention. Based on Cohen’s (1988) guidelines, Attitude (f² = 0.099) demonstrates a medium effect, indicating it has the most substantial influence on intention. Perceived Behavioral Control (f² = 0.071) shows a small-to-medium effect, while Knowledge (f² = 0.055), Subjective Norm (f² = 0.044), and Health Need (f² = 0.031) exhibit small effects. Despite varying magnitudes, all predictors contribute meaningfully to the model’s explanatory power.
Table 10: Effect Size (f²)
Path | f² Effect Size | Interpretation |
HN → INT | 0.031 | Small |
KN → INT | 0.055 | Small |
ATT → INT | 0.099 | Medium |
SN → INT | 0.044 | Small |
PBC → INT | 0.071 | Small–Medium |
The structural model results demonstrate strong explanatory and predictive power. All five predictors significantly influence intention to consume functional food, with Attitude showing the strongest effect. The R² value of 0.612 indicates substantial variance explained, while Q² = 0.372 confirms high predictive relevance. No multicollinearity issues were detected (VIF < 5). Effect size analysis shows that all constructs contribute meaningfully, with Attitude and Perceived Behavioral Control having the most notable impacts. These findings support the model’s robustness and theoretical soundness in explaining consumer intention.
DISCUSSION
To examine the influence of health needs, knowledge, attitude, subjective norms, and perceived behavioral control on the intention to consume functional food among senior citizens in Sabah.
The structural model results confirm that all five variables, health needs, knowledge, attitude, subjective norms, and perceived behavioral control, significantly influence senior citizens’ intention to consume functional food. The strongest predictor was attitude (β = 0.275, p < 0.001), followed by perceived behavioral control and knowledge. These findings are consistent with the Theory of Planned Behavior (TPB), which posits that behavioral intention is primarily shaped by an individual’s attitude, perceived control, and normative beliefs (Ajzen, 1991). Health need, though statistically significant, showed a modest effect size. This suggests that while health concerns motivate seniors, they do not act in isolation without positive attitudes or sufficient knowledge. This aligns with previous studies which found that perceived health benefits increase interest in functional food, but are often mediated by knowledge and accessibility (Ares & Gámbaro, 2007; Verbeke, 2006).
Subjective norms also played a notable role, indicating that family, caregivers, or peers influence dietary behavior in later life, a finding supported by Ko et al. (2013), who reported similar patterns among aging Korean consumers. Implications include designing targeted educational campaigns and social marketing strategies that enhance seniors’ attitudes and control beliefs. Public health initiatives should also focus on strengthening community and family-based advocacy to create normative support.
To assess the most significant psychological and social predictors of functional food consumption intention within the Theory of Planned Behavior framework.
This study reaffirms the Theory of Planned Behavior (TPB) as an appropriate model for understanding senior citizens’ functional food consumption. Among the TPB constructs, attitude emerged as the most influential predictor (β = 0.275), followed by perceived behavioral control (PBC) (β = 0.231) and subjective norms (β = 0.183). These findings echo previous research where attitude consistently outperformed other TPB constructs in predicting health-related behavior (Chen, 2011; Dean et al., 2012). The prominence of attitude reflects how personal beliefs about the health-enhancing properties of functional food, such as cholesterol reduction, immunity boosting, or improved digestion, shape consumption behavior. This supports findings by Siro et al. (2008) that consumers perceive functional foods positively when they understand their physiological benefits.
Perceived behavioral control was also influential, suggesting that seniors’ intention depends on their confidence in accessing, affording, and incorporating such products into their diet. Accessibility, pricing, and availability in local markets likely play a role here, reinforcing findings from Bech-Larsen & Grunert (2003). Subjective norms, while the least dominant among the TPB constructs, still had a meaningful effect. This implies that seniors are influenced by their caregivers, spouses, and health practitioners, especially in collectivist societies like Malaysia where familial ties are strong (Teng & Wang, 2015). These insights support public health messaging and behavior change campaigns that build self-efficacy, correct misconceptions, and normalize functional food consumption within family and community networks.
To explore how demographic characteristics (e.g., age, education level, income, and residential area) relate to the intention to consume functional food among the elderly population in Sabah.
The demographic analysis reveals important contextual factors that may influence consumption intention. The majority of respondents were aged 60–69, female, and from lower income brackets, characteristics common among Malaysia’s elderly population (DOSM, 2022). Interestingly, respondents with higher education levels and urban residency were more likely to report a stronger intention to consume functional food, likely due to better health literacy and access to diversified food options (Annunziata & Vecchio, 2013). While age was not found to be a direct predictor in the structural model, it potentially moderates other relationships, as older seniors (>70) may experience reduced autonomy or dietary conservatism.
Income also plays a crucial role. Lower-income seniors, despite high health needs, may struggle to afford functional foods, indicating the importance of affordability and subsidy-based policy interventions. These results highlight the need for targeted outreach strategies that consider socioeconomic and geographic disparities. For example, educational campaigns and retail initiatives should be tailored for rural or low-income communities through subsidized health programs or community-based food distribution models. From a policy perspective, the findings support the inclusion of functional food education in government health screenings or senior welfare programs (MOH, 2021). For industry stakeholders, there is potential to expand affordable functional food lines tailored to senior palates and nutritional needs.
CONCLUSION
This study investigated the factors influencing senior citizens’ intention to consume functional food in Sabah, Malaysia, guided by the Theory of Planned Behavior. The findings confirmed that health needs, knowledge, attitude, subjective norms, and perceived behavioral control significantly shape consumption intention, with attitude emerging as the strongest predictor. The model explained 61.2% of the variance in intention, indicating robust explanatory power. The study contributes theoretically by validating TPB in the context of elderly nutrition behavior and extends practical insights for health promotion and industry stakeholders in developing targeted interventions.
From a policy perspective, the results underscore the need for educational outreach, pricing strategies, and accessibility improvements, particularly for low-income and rural seniors. However, the study is limited by its cross-sectional design and regional focus, which may affect generalizability. Future research should explore longitudinal impacts, behavioral outcomes beyond intention, and include qualitative methods to capture deeper motivations and barriers. Additionally, comparative studies across Malaysian states or ASEAN regions would enhance understanding of cultural and environmental influences on functional food consumption. Expanding this research will further inform the development of inclusive, health-oriented food policies for aging populations.
REFERENCES
- Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179–211.
- Annunziata, A., & Vecchio, R. (2013). Consumer perception of functional foods: A conjoint analysis with probiotics. Food Quality and Preference, 28(1), 348–355. https://doi.org/10.1016/j.foodqual.2012.10.009
- Annunziata, A., & Vecchio, R. (2019). Consumer Behaviour towards functional foods: A review. International Journal on Food System Dynamics, 10(4), 280–294. https://doi.org/10.18461/ijfsd.v10i4.1047
- Ares, G., & Gámbaro, A. (2007). Influence of gender, age and motives underlying food choice on perceived healthiness and willingness to try functional foods. Appetite, 49(1), 148–158. https://doi.org/10.1016/j.appet.2007.01.006
- Armitage, C. J., & Conner, M. (2001). Efficacy of the Theory of Planned Behaviour: A meta‐analytic review. British Journal of Social Psychology, 40(4), 471–499.
- Bech-Larsen, T., & Grunert, K. G. (2003). The perceived healthiness of functional foods: A conjoint study of Danish, Finnish and American consumers’ perception. Appetite, 40(1), 9–14. https://doi.org/10.1016/S0195-6663(02)00171-X
- Bentler, P. M., & Bonett, D. G. (1980). Significance tests and goodness of fit in the analysis of covariance structures. Psychological Bulletin, 88(3), 588–606.
- Chen, M. F. (2011). The joint moderating effect of health consciousness and healthy lifestyle on consumers’ willingness to use functional foods in Taiwan. Appetite, 57(1), 253–262.
- Dean, M., Rakats, M. M., & Shepherd, R. (2012). Moral concerns and consumer choice of conventional and genetically modified food. Appetite, 59(2), 343–349. https://doi.org/10.1016/j.appet.2012.05.024
- Dean, M., Shepherd, R., & Arvola, A. (2012). Consumer perceptions of healthy food and dietary choice. In Frewer, L. (Ed.), Understanding Consumers of Food Products (pp. 95–120). Woodhead Publishing.
- Department of Statistics Malaysia (DOSM). (2022). Population and Demographic Indicators, Malaysia. https://www.dosm.gov.my
- Fishbein, M., & Ajzen, I. (2010). Predicting and changing behavior: The reasoned action approach. Psychology Press.
- Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39–50.
- Granato, D., Nunes, D. S., & Barba, F. J. (2020). An integrated strategy between food chemistry, biology, nutrition, pharmacology, and statistics in the development of functional foods. Trends in Food Science & Technology, 95, 128–137. https://doi.org/10.1016/j.tifs.2019.11.010
- Hair, J. F., Hult, G. T. M., Ringle, C. M., & Sarstedt, M. (2019). A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM) (2nd ed.). Sage Publications.
- Hair, J. F., Matthews, L. M., Matthews, R. L., & Sarstedt, M. (2021). PLS-SEM or CB-SEM: updated guidelines on which method to use. International Journal of Multivariate Data Analysis, 5(3), 1–22.
- Hasler, C. M. (2000). The changing face of functional foods. The Journal of the American College of Nutrition, 19(5), 499S–506S. https://doi.org/10.1080/07315724.2000.10718975
- Henseler, J., Hubona, G., & Ray, P. A. (2016). Using PLS path modeling in new technology research: Updated guidelines. Industrial Management & Data Systems, 116(1), 2–20.
- Henseler, J., Ringle, C. M., & Sarstedt, M. (2015). A new criterion for assessing discriminant validity in variance-based structural equation modeling. Journal of the Academy of Marketing Science, 43(1), 115–135.
- Kamarulzaman, N. H., Zainuddin, Y., & Hashim, N. H. (2020). Consumers’ willingness to pay for functional foods: Evidence from Malaysia. Journal of Agribusiness Marketing, 13, 43–55. https://doi.org/10.22004/ag.econ.303267
- Ko, L. E., So, J., & Lee, Y. (2013). Aging, health, and food: Food-related beliefs and practices among Korean older adults. Journal of Nutrition Education and Behavior, 45(3), 173–180. https://doi.org/10.1016/j.jneb.2012.10.001
- Lee, Y. H., Rahman, A. A., & Abdul Rashid, N. R. N. (2022). Factors affecting older adults’ healthy food choice: A Malaysian perspective. Malaysian Journal of Consumer and Family Economics, 28(S1), 68–83.
- Madli, F., Janin, Y., Gulabdin, S., Lada, S., Yun, W. S., Adis, A. A. A., & Jafar, A. (2024a). Artificial Intelligence and Public Health Context: What We Should Know?. Journal of Advanced Research in Applied Sciences and Engineering Technology, 39(2), 96-109.
- Madli, F., Sondoh, S., Totu, A., T, R., Janin, Y., Syed Annuar, S. N., & Cham, T. H. (2024b). Modelling organ donation information adoption among Malaysian youth using the information adoption model (IAM). International Journal of Pharmaceutical and Healthcare Marketing, 18(2), 252-275.
- Ministry of Health Malaysia (MOH). (2021). National Health and Morbidity Survey (NHMS). https://iku.moh.gov.my
- Ministry of Health Malaysia. (2021). National Health and Morbidity Survey (NHMS). https://iku.moh.gov.my.
- Niva, M. (2007). “All foods affect health”: Understandings of functional foods and healthy eating among health-oriented Finns. Appetite, 48(3), 384–393.
- Norazlanshah, H., Karim, R., Norliana, I., et al. (2021). Functional food awareness among Malaysian adults. Journal of Functional Foods, 85, 104641. https://doi.org/10.1016/j.jff.2021.104641
- Pappalardo, G., & La Barbera, F. (2018). Using the Theory of Planned Behavior to predict parents’ intention to purchase functional foods for their children. Food Quality and Preference, 72, 1–6. https://doi.org/10.1016/j.foodqual.2018.09.001
- Sarstedt, M., Ringle, C. M., & Hair, J. F. (2019). PLS-SEM: Looking back and moving forward. Long Range Planning, 52(1), 105-111.
- Siro, I., Kapolna, E., Kapolna, B., & Lugasi, A. (2008). Functional food. Food Research International, 41(3), 457–465. https://doi.org/10.1016/j.foodres.2008.03.005
- Talwar, M., Talwar, S., Tandon, A., & Kaur, P. (2021). Factors influencing consumer purchase intentions for functional foods: A meta-analytic review. Journal of Consumer Behaviour, 20(6), 1481–1501. https://doi.org/10.1002/cb.1946
- Teng, C. C., & Wang, Y. M. (2015). Decisional factors driving organic food consumption: Generation of consumer purchase intentions. British Food Journal, 117(3), 1066–1081. https://doi.org/10.1108/BFJ-12-2013-0361
- United Nations. (2023). World Population Prospects 2023. https://population.un.org/wpp/
- Verbeke, W. (2006). Functional foods: Consumer willingness to compromise on taste for health? Food Quality and Preference, 17(1–2), 126–131. https://doi.org/10.1016/j.foodqual.2005.03.003