Attitudes of undergraduate nursing students toward caring for older adults: a cross-sectional study from Limpopo Province, South Africa Authors: Tshepo A Ntho, Linda Nchabeleng, Thabo A Phukubye, Peter M Mphekgwana Published: 2026 ABSTRACT AbstractBackgroundAttitudes of nursing students toward older adults are a critical determinant of the quality of geriatric care they will provide upon graduation. In South Africa, nursing education does not sufficiently incorporate dedicated gerontological nursing content. This study aimed to assess the attitudes of undergraduate nursing students toward caring for older adults within a South African nursing training context.MethodsA quantitative cross-sectional study was conducted among 246 undergraduate nursing students at a historically disadvantaged institution in Limpopo Province, South Africa, using stratified random sampling. Data were collected online via Google Forms using Kogan's Attitudes toward Old People scale (KAOP) and a demographic questionnaire. Data were analysed using descriptive statistics, independent samples t-test, one-way ANOVA, chi-square test, and binary logistic regression using SPSS version 29.0.ResultsA total of 246 undergraduate nursing students completed the survey. The majority of students (54.9%) demonstrated slightly positive attitudes toward older adults, with a mean KAOP total score of 142.70 (SD = 13.02). Age group and gender were identified as significant independent predictors of attitudes. Compared with students aged 18-20 years, those aged 21-23 years (AOR = 0.361, 95% CI = 0.201–0.648, p < 0.001) and 24–26 years (AOR = 0.171, 95% CI = 0.070–0.419, p < 0.001) had significantly lower odds of demonstrating strongly positive attitudes toward older adults. Male students demonstrated higher odds of holding strongly positive attitudes compared to female students (AOR = 0.438, 95% CI = 0.249–0.772, p = 0.004). Year of study, previous experience with older adults, and intention to work with older adults were not significantly associated with attitudes (all p > 0.05).ConclusionPositive attitudes toward older adult care are not uniformly held among undergraduate nursing students enrolled in a South African nursing education programme, and attitudinal formation is driven by age and gender rather than academic progression or clinical exposure alone. Systematic integration of gerontological nursing content within the R174 curriculum is recommended to address attitudinal deficits and promote equitable geriatric care in South Africa.Keywords: Older adults, attitudes, undergraduate nursing students, gerontological nursing, South Africa Introduction and background The global demographic landscape is undergoing a profound transformation. The World Health Organization projects that the number of people aged 60 years and older will double by 2050 [1]. As the fastest-growing population group, older adults are placing increasing demands on health systems worldwide [2, 3]. Older adults are defined as individuals aged 60 years and above, consistent with the South African Older Persons Act and the World Health Organization’s definition of older persons in the African context [1, 4]. Ageing is a natural, universal process that introduces complex physical, psychological, and socio-economic needs, requiring healthcare providers who are not only clinically competent but also equipped with positive attitudes toward older adults. Attitudes held by nursing students toward older adults are particularly significant, as these attitudes shape the quality of care delivered and influence career choices in gerontological nursing [5, 6]. Given the rapid growth of the older adult population, negative attitudes among nursing students may persist into professional practice and adversely affect the quality of health care provided to older persons who increasingly rely on healthcare services [7, 8]. Importantly, as this population grows, the demand for compassionate, attitude-aligned gerontological nursing increases. However, negative attitudes toward older adults remain a persistent concern globally. The literature suggests that nursing students’ attitudes toward older adults are complex and vary across educational and cultural contexts. For instance, a recent meta-analysis involving more than 24,000 nursing students from 11 countries found that attitudes toward older adults ranged from neutral to moderately positive [9]. In Portugal, negative and unattractive representations of social problems associated with ageing appear to shape nursing students’ attitudes toward older people [10]. In contrast, a study conducted in Spain demonstrated that nursing students generally held positive attitudes toward caring for older adults [11]. Within Africa, evidence regarding nursing students’ attitudes toward older adults is similarly mixed. In Egypt, over 80% of nursing students held negative views on ageing-related physiological changes [12]. In contrast, in Nigeria, nursing students demonstrated positive attitudes toward elderly care [5]. Across Sub-Saharan Africa, attitudinal deficits may be compounded by insufficient geriatric content in nursing curricula. For instance, Ghana’s nursing curriculum has been reported to provide limited caregiving content related to older adults, potentially restricting opportunities for students to develop positive engagement with this population [13]. These findings highlight the influence of educational preparation on nursing students’ attitudes toward older adults across Africa. Evidence consistently shows that students who complete geriatric coursework or clinical placements develop more positive attitudes toward older adults [11, 14]. Work-integrated learning (WIL), an educational approach that integrates classroom learning with supervised clinical practice, is a key strategy in this regard. Through WIL, nursing students assess, monitor, and manage patients under supervision, cultivating empathy and professional disposition [15, 16]. However, exposure to residential aged-care settings remains limited. International evidence indicates that nursing students often have limited prior knowledge of residential aged care and insufficient exposure to care homes, which may affect their preparedness and confidence in caring for older adults [17, 18]. However, exposure to residential aged-care settings remains limited. International evidence indicates that nursing students often have limited prior knowledge of residential aged care and insufficient exposure to care homes, which may affect their preparedness and confidence in caring for older adults [17, 18]. In South Africa, old age homes have not been consistently utilised as clinical learning sites for nursing students, and concerns have been raised regarding their readiness to support clinical training [16]. These challenges may limit opportunities for meaningful engagement with older adults during training and the development of positive attitudes towards their care. Despite ongoing nursing education reforms aimed at aligning training with global standards, the preparation of nursing students for gerontological care remains inconsistent. Studies conducted among nursing students and other health professions students have reported varying attitudes toward older adults, highlighting inconsistencies in the development of positive attitudes towards ageing during professional training [19–21]. Nevertheless, a study conducted in the Western Cape reported generally positive attitudes among nursing students toward older adults [22]. These contrasting findings suggest that attitude formation is not uniformly addressed across provinces. Similar concerns have been raised regarding undergraduate nursing education, where gerontological nursing receives limited emphasis within nursing and other health professions training programmes [19, 22–24]. The limited focus on geriatric care training is particularly concerning, given the rapid growth of the population aged over 60 years [23, 25, 26]. The Bachelor of Nursing curriculum places limited clinical emphasis on geriatric care, which directly influences how students interact with and perceive older adult patients [24]. Furthermore, geriatric content remains insufficiently addressed in South African nursing and medical training programmes, which may have downstream effects on students’ attitudes toward older adults [23]. Subsequently, old age homes have not been consistently utilised as clinical learning sites for nursing students in South Africa, and concerns have been raised regarding their readiness to support clinical training [16]. These challenges may limit opportunities for meaningful engagement with older adults during training and the development of positive attitudes towards their care. Despite growing recognition of the importance of gerontological nursing education in South Africa, evidence regarding nursing students’ attitudes toward older adults in Limpopo Province remains limited. This evidence gap constrains efforts to develop contextually relevant educational strategies to improve the quality of care for older persons. Against this backdrop, this study assessed nursing students’ attitudes toward caring for older adults in Limpopo Province, using the KOAP Scale [27]. Understanding student attitudes in Limpopo Province contributes to evidence-based educational reform and supports policy decisions aimed at strengthening geriatric nursing education in South Africa. The findings may further inform curriculum development and promote the meaningful integration of gerontological principles into nursing training. Problem statement Nursing students are expected to develop positive attitudes toward older adults during their training, as such attitudes are essential for providing compassionate, equitable, and person-centred geriatric care. Positive attitudes have been associated with improved communication, enhanced quality of care, and greater willingness to work with older adult populations [2, 8, 22]. However, negative attitudes toward older adults remain prevalent among nursing students globally, posing a significant challenge to the delivery of quality geriatric care by the future nursing workforce [10, 23, 28]. Negative attitudes toward older adults may result in ageist behaviours, reduced quality of care, ineffective communication, and reluctance to engage in gerontological nursing practice. Such attitudes are particularly concerning given the rapidly growing older adult population and their increasing healthcare needs. Although several factors may contribute to these attitudes, previous studies have highlighted limited gerontological preparation and inadequate exposure to older adult care during training as potential influences [16, 19, 23, 24]. Despite growing evidence from other settings, empirical evidence on undergraduate nursing students’ attitudes toward older adults in Limpopo Province remains absent. This lack of evidence limits the ability of Nursing Education Institutions (NEIs) and policymakers to develop contextually relevant educational strategies to promote positive attitudes toward older adults and strengthen the quality of geriatric care. Research question What are the attitudes of undergraduate nursing students toward older adults, and which demographic characteristics are associated with and predict these attitudes? Objectives To assess the attitudes of undergraduate nursing students toward older adults using the KOAP scale. To assess the attitudes of undergraduate nursing students toward older adults using the KOAP scale. To determine the association between demographic characteristics and nursing students’ attitudes toward older adults and identify predictors of positive attitudes. To determine the association between demographic characteristics and nursing students’ attitudes toward older adults and identify predictors of positive attitudes. Methods This study employed a quantitative, cross-sectional design to assess the attitudes of undergraduate nursing students toward caring for older adults. A cross-sectional design was appropriate for measuring attitudes and their associated factors at a single point in time within a defined population [29, 30]. This study adhered to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines to ensure rigorous and transparent reporting of the study findings [31]. Setting and population The study was conducted at a selected Nursing Education Institution (NEI) in Limpopo Province, South Africa. The NEI is one of eight higher education institutions in the country formally recognised as historically disadvantaged and is accredited by the South African Nursing Council (SANC) to offer the Bachelor of Nursing Science programme in accordance with Regulation R174 [32]. The institution was selected as the study site because it provided access to the target population and served as the setting within which the undergraduate nursing research project was conducted. The Bachelor of Nursing Science programme integrates theoretical instruction with WIL, which is undertaken at accredited healthcare facilities across the province. The target population comprised all undergraduate nursing students enrolled across the four years of the Bachelor of Nursing Science programme and registered with SANC as learner nurses in accordance with the Nursing Act No. 33 of 2005 [33]. The inclusion of students from all levels of study enabled a comprehensive assessment of attitudes toward older adults across different stages of nursing education and varying levels of clinical exposure. This approach facilitated the examination of attitudinal patterns among students with differing levels of academic preparation and practical experience in caring for older adults. Sample size and sampling The study population comprised 303 undergraduate nursing students registered across all four academic levels of the Bachelor of Nursing Science programme at the selected NEI in Limpopo Province, South Africa, distributed as follows: Level 1 (n = 80), Level 2 (n = 78), Level 3 (n = 72), and Level 4 (n = 73). A stratified sampling approach was employed to ensure adequate and proportional representation across all academic levels. The total population was stratified by academic level, with each stratum treated as an independent population for the purpose of sample size calculation. The required sample size for each stratum was calculated separately using the Raosoft sample size calculator, applying a margin of error of 5% and a confidence level of 95% [34]. This yielded the following stratum-specific sample sizes: Level 1 (n = 53), Level 2 (n = 62), Level 3 (n = 64), and Level 4 (n = 67), resulting in a total sample of 246 participants. Convenience sampling was subsequently employed within each stratum [29, 30]. The electronic questionnaire was distributed through the official WhatsApp groups of the respective academic levels, and eligible students who met the inclusion criteria were invited to participate in the study. Participants were eligible for inclusion if they were registered in the Bachelor of Nursing Science programme at the selected NEI, enrolled in any of the four academic levels, had been exposed to clinical practice in both hospital and primary healthcare settings, and voluntarily provided informed consent prior to participation. All 246 participants completed the questionnaire in full, yielding a response rate of 100% with no missing data. Measurement The first section comprised a researcher-developed questionnaire that collected participants’ demographic characteristics, including age, gender, race, level of study, previous experience working with older adults, and intention to work with older adults upon graduation, using categorical response options. These variables provided essential contextual information for examining associations between demographic factors and attitudes toward older adults. The second section measured participants’ attitudes toward older adults using the KAOP scale, originally developed by Kogan [27]. The KAOP scale was adopted in its original published English form and used without translation, cultural adaptation, or modification. The instrument has been widely used and validated among nursing students and other health professions students in South Africa and internationally [2, 13, 22, 35]. The KAOP scale was adopted from the original published source and has demonstrated utility and validity across diverse nursing education and healthcare contexts. Importantly, before the main study, the questionnaire was pilot tested among 25 undergraduate nursing students to assess the clarity, relevance, and comprehensibility of the questionnaire items, as well as the functionality of the electronic data collection process. The pilot participants possessed characteristics similar to those of the study population but were excluded from the main study. Feedback obtained during pilot testing confirmed the suitability of the questionnaire for administration in the main study. Data collected during the pilot test were not included in the final analysis. The scale consists of 34 items arranged in paired positively worded (KAOP+) and negatively worded (KAOP−) statements, with 17 items in each subscale. Responses are rated on a six-point Likert scale ranging from 1 (Strongly Disagree) to 6 (Strongly Agree), as recommended by the original author [27]. All negatively worded items were reverse coded before analysis so that higher total scores consistently reflected more positive attitudes toward older adults, while lower scores indicated more negative perceptions. The theoretical total score ranges from 34 to 204, with a midpoint of 119 representing a neutral attitudinal position [27]. Data collection Following the class visits, a study information leaflet containing the study purpose, objectives, eligibility criteria, and a link to the online questionnaire was distributed through the official WhatsApp groups of each academic level. Eligible students were invited to participate voluntarily in the study. To enhance participation, reminder messages and study leaflets were shared through the WhatsApp groups at regular intervals throughout the data collection period. Recruitment continued until the required sample size for each academic level had been achieved. The questionnaire was administered electronically using Google Forms and commenced with an electronic informed consent form. Participants were required to read the consent information and indicate their willingness to participate electronically before accessing the questionnaire items. Only participants who provided electronic consent were able to proceed to the questionnaire. The online mode of administration facilitated convenient access to the questionnaire while minimising disruption to students’ academic activities. Completion of the questionnaire required approximately 20 to 25 min. Participation was entirely voluntary, and to ensure anonymity, no personally identifying information was collected, and responses were submitted anonymously through the online platform. Data analysis Collected data were analysed using the Statistical Package for the Social Sciences (SPSS) version 29.0. Descriptive statistics, including frequencies, percentages, means, and standard deviations, were computed to summarise participants’ sociodemographic characteristics and KAOP scores. The internal consistency of the KAOP scale was assessed using Cronbach’s alpha coefficient. For the primary analysis, continuous KAOP scores were used to examine group differences across sociodemographic variables. Independent samples t-tests were used to compare KAOP scores by gender, while one-way analysis of variance (ANOVA) was applied to examine differences across age group, year of study, previous experience with older adults, and intention to work with older adults upon graduation. The homogeneity of variance assumption was assessed using Levene’s test, with Games-Howell post hoc comparisons applied where the assumption was violated, and Tukey’s HSD applied where variances were equal. The effect size for statistically significant ANOVA results was reported using eta-squared (η²). For the secondary analysis, KAOP total scores were dichotomised using the sample mean of 142.70 as the cut-off point, whereby scores at or below the mean were classified as slightly positive attitudes and scores above the mean were classified as strongly positive attitudes. This approach is consistent with recommendations for the use of mean-based dichotomisation in culturally diverse populations, where the theoretical midpoint may yield markedly unequal distributions [13]. Chi-square tests were conducted to examine associations between sociodemographic variables and binary attitude classification, with Cramer’s V reported as the effect size measure. Binary logistic regression was subsequently performed to identify independent predictors of attitude classification, with unadjusted crude odds ratios (COR) and adjusted odds ratios (AOR) reported alongside 95% confidence intervals. Model fit was evaluated using the Hosmer-Lemeshow goodness-of-fit test and Nagelkerke R². A p-value of less than 0.05 was considered statistically significant for all analyses. Ethical considerations Ethical clearance for this study was granted by the Turfloop Research Ethics Committee (TREC ref: TREC 1447/2024:UG). Institutional permission to access and engage with undergraduate nursing students was obtained from the School of Healthcare Sciences Director and the Head of the Department of Nursing Science at the selected NEI. The study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki, ensuring the protection, rights, and well-being of all participants throughout the research process. Prior to data collection, the research team visited each class across all four academic levels to provide a thorough verbal explanation of the study’s purpose, procedures, potential risks, and benefits, and to address any questions raised by prospective participants. Following this explanation, all prospective participants received a written informed consent form and were required to provide voluntary consent before being enrolled in the study. Students were explicitly advised that participation was entirely voluntary and that withdrawal at any point would carry no negative consequences for their academic standing or relationship with the institution. Anonymity was maintained throughout by refraining from collecting any personally identifying information from respondents. All data were handled with strict confidentiality and stored securely in accordance with institutional data protection requirements. Results Sociodemographic characteristics of nursing students A total of 246 undergraduate nursing students participated in the study, achieving a 100% response rate. The majority of participants were female (66.7%), with males comprising one-third of the sample (33.3%). More than half (51.6%) were aged between 21 and 23 years, followed by those aged 18 to 20 years (33.3%) and 24 to 26 years (15.0%). The sample was distributed across all four academic levels, ranging from 21.5% (Level 1) to 27.2% (Level 4). All participants identified as Black African, which is consistent with the predominantly Black African student demographic characteristic of historically disadvantaged universities in South Africa [36]. Regarding previous experience working with older people, 48.0% reported prior experience, 29.7% reported none, and 22.4% were uncertain. Concerning intention to work with older people upon graduation, 50.4% were uncertain, 38.6% indicated an intention to do so, and 11.0% did not. The sociodemographic characteristics of the participants are presented in Table 1. Table 1Sociodemographic Characteristics of Participants (N = 246)Variable n %Gender Male8233.3 Female16466.7Age Group 18–20 years8233.3 21–23 years12751.6 24–26 years3715.0Year of Study Level 15321.5 Level 26225.2 Level 36426.0 Level 46727.2Previous Experience Working with Older People Yes11848.0 No7329.7 Uncertain5522.4Intention to Work with Older People upon Graduation Yes9538.6 No2711.0 Uncertain12450.4All participants identified as Black African; race was therefore excluded from inferential analysis Sociodemographic Characteristics of Participants (N = 246) All participants identified as Black African; race was therefore excluded from inferential analysis Reliability of the KAOP Scale The internal consistency of the 34-item KAOP scale was assessed using Cronbach’s alpha, yielding a coefficient of 0.663, which is considered moderate to reasonable according to Taber [37]. While marginally below the conventionally accepted threshold of 0.70, this value is consistent with reliability estimates reported for the full 34-item KAOPS across diverse nursing student populations internationally. Among Ghanaian nursing students, a Cronbach’s alpha of 0.65 was reported only after the removal of 10 items with low or negative corrected item-to-total correlations, with no further improvement observed following additional item removal [13]. Similarly, a multi-country validation study among nursing students in Thailand, Myanmar, and Indonesia reported total scale alpha coefficients of 0.70, 0.68, and 0.66, respectively, for the full 34-item scale, with all items demonstrating significant item-to-total correlations across all three countries [38]. Collectively, these findings suggest that moderate alpha coefficients are characteristic of the full 34-item KAOPS across diverse cultural contexts, and that values in the range of 0.63–0.70 represent an acceptable level of internal consistency for this instrument. The full 34-item scale was therefore retained to preserve the integrity of the paired subscale structure and to facilitate direct comparability with the international nursing student literature. Attitudes of undergraduate nursing students toward older adults The mean KAOP Total score was 142.70 (SD = 13.02), with observed scores ranging from 118 to 194 out of a theoretical maximum of 204 and a theoretical midpoint of 119 [27]. Almost all participants scored above the theoretical midpoint, indicating that the sample as a whole held positive attitudes toward older adults. The KAOP scale has previously been classified into six attitudinal categories ranging from strongly negative to strongly positive [39]. Based on this classification, the overall sample demonstrated slightly positive attitudes toward older adults, with a mean score of 142.70 falling within the slightly positive range (136–169). The mean Negative Attitude Subscale (KAOP−) score was 72.73 (SD = 8.15), exceeding the subscale midpoint of 59.5, indicating a tendency to reject negative stereotypes about older adults. Similarly, the mean Positive Attitude Subscale (KAOP+) score was 69.96 (SD = 7.46), also above the subscale midpoint of 59.5, reflecting a general endorsement of positive attributes of older adults. Notably, the KAOP− mean exceeded the KAOP+ mean, suggesting that nursing students in this study were more proficient at rejecting negative stereotypes than at actively affirming the positive characteristics of older adults. Descriptive statistics for the three KAOP scores are presented in Table 2. Table 2Descriptive Statistics for KAOP Scores(N = 246)ScaleMeanSDMinMaxαKAOP Total (34 items; range 34–204)142.7013.021181940.663KAOP− Negative Subscale (17 items; range 17–102)72.738.1554102—KAOP+ Positive Subscale (17 items; range 17–102)69.967.464697—α = Cronbach’s alpha reported for the total scale only. KAOP + = Positive Attitude Subscale; KAOP− = Negative Attitude Subscale. Subscale midpoint = 59.5; Total scale theoretical midpoint = 119 Descriptive Statistics for KAOP Scores(N = 246) α = Cronbach’s alpha reported for the total scale only. KAOP + = Positive Attitude Subscale; KAOP− = Negative Attitude Subscale. Subscale midpoint = 59.5; Total scale theoretical midpoint = 119 Association between demographic characteristics and attitudes Independent samples t-tests revealed no statistically significant differences between male and female participants on any of the three continuous KAOP scores. Although males scored marginally higher than females on all three outcomes (KAOP Total: 144.70 ± 14.30 vs. 141.70 ± 12.26; KAOP+: 70.73 ± 8.11 vs. 69.58 ± 7.10; KAOP−: 73.96 ± 8.64 vs. 72.12 ± 7.85), none of these differences reached statistical significance (p = 0.089, p = 0.254, and p = 0.094, respectively). One-way ANOVA revealed a statistically significant effect of age group on all three continuous KAOP scores. The youngest age group (18–20 years) demonstrated the highest mean scores across all three outcomes (KAOP Total: 149.34 ± 17.13; KAOP+: 72.76 ± 9.17; KAOP−: 76.59 ± 9.99), with scores declining progressively with increasing age. The oldest age group (24–26 years) recorded the lowest means (KAOP Total: 137.11 ± 8.49; KAOP+: 67.35 ± 6.25; KAOP−: 69.76 ± 6.14). The ANOVA was statistically significant for all three outcomes: KAOP Total (F = 19.228, p < 0.001, η²=0.137); KAOP− (F = 15.863, p < 0.001); and KAOP+ (F = 9.937, p < 0.001). The eta-squared value of 0.137 indicates a moderate effect size for KAOP Total. Post hoc Games-Howell comparisons confirmed significant differences between the youngest age group and both older age groups (all p < 0.001), whereas no significant difference was observed between the 21–23 and 24-26-year groups. Year of study was not significantly associated with attitudes toward older adults across any of the three KAOP scores (all p > 0.05). This implies that academic progression alone does not drive attitude change in this sample. Previous experience working with older people was not significantly associated with any KAOP score at the omnibus level (KAOP Total: F = 3.002, p = 0.052; KAOP−: F = 2.195, p = 0.114; KAOP+: F = 1.989, p = 0.139). No significant association was found between intention to work with older people upon graduation and any of the three continuous KAOP scores (all p > 0.05). These findings are presented in Table 3. Table 3Association Between Sociodemographic Variables and Continuous KAOP Scores(N = 246)VariableCategoryKAOP Total Mean ± SDKAOP+ Mean ± SDKAOP− Mean ± SDGender (Independent Samples t-test)Male144.70 ± 14.3070.73 ± 8.1173.96 ± 8.64Female141.70 ± 12.2669.58 ± 7.1072.12 ± 7.85p-value0.0890.2540.094Age Group (One-Way ANOVA)18–20 years149.34 ± 17.1372.76 ± 9.1776.59 ± 9.9921–23 years140.03 ± 8.6968.92 ± 5.9071.11 ± 6.2724–26 years137.11 ± 8.4967.35 ± 6.2569.76 ± 6.14p-value< 0.001**< 0.001**< 0.001**Year of Study (One-Way ANOVA)Level 1142.45 ± 13.7870.17 ± 7.3272.28 ± 8.42Level 2140.05 ± 9.9468.16 ± 6.7771.89 ± 6.88Level 3143.23 ± 14.6669.98 ± 8.0373.25 ± 9.07Level 4144.82 ± 13.1171.45 ± 7.4173.37 ± 8.18p-value0.2150.0960.685Previous Experience Working with Older People (One-Way ANOVA)Yes143.86 ± 13.8070.47 ± 7.8673.39 ± 8.35No141.12 ± 10.6369.38 ± 6.6671.73 ± 7.24Uncertain136.65 ± 10.8366.94 ± 5.0169.71 ± 8.56p-value0.0520.1390.114Intention to Work with Older People upon Graduation (One-Way ANOVA)Yes143.22 ± 13.7170.37 ± 7.4972.85 ± 8.72No143.11 ± 14.0370.29 ± 8.2872.82 ± 8.36Uncertain141.24 ± 10.4468.84 ± 6.4672.40 ± 6.79p-value0.5980.3900.935KAOP Total = Total KAOP score; KAOP + = Positive Attitude Subscale; KAOP− = Negative Attitude Subscale. Values = Mean ± SD. Post hoc: Games-Howell, where Levene’s p < 0.05; Tukey HSD where variances were equal. **p < 0.001 Association Between Sociodemographic Variables and Continuous KAOP Scores(N = 246) KAOP Total = Total KAOP score; KAOP + = Positive Attitude Subscale; KAOP− = Negative Attitude Subscale. Values = Mean ± SD. Post hoc: Games-Howell, where Levene’s p < 0.05; Tukey HSD where variances were equal. **p < 0.001 Binary attitude classification Given that almost all participants scored above the theoretical midpoint of 119, rendering binary classification based on the theoretical midpoint analytically meaningless. The sample mean of 142.70 was used as the cut-off for dichotomisation, consistent with recommendations for culturally diverse samples, where the theoretical midpoint produces severely unequal distributions [13]. Participants scoring at or below 142.70 were classified as having slightly positive attitudes (n = 135; 54.9%), while those scoring above 142.70 were classified as having strongly positive attitudes (n = 111; 45.1%), as presented in Table 4. Table 4Distribution of Binary Attitude Categories (N = 246)Category n %Slightly Positive Attitudes (score ≤ 142.70)13554.9Strongly Positive Attitudes (score > 142.70)11145.1Total246100.0Cut-off based on sample mean (142.70) as recommended for culturally diverse samples (Abudu-Birresborn et al., 2023). Classification based on Rathnayake et al. (2016) Distribution of Binary Attitude Categories (N = 246) Cut-off based on sample mean (142.70) as recommended for culturally diverse samples (Abudu-Birresborn et al., 2023). Classification based on Rathnayake et al. (2016) Secondary analysis - Chi-square and logistic regression results Chi-square analysis revealed that gender (χ²=7.387, df = 1, p = 0.007; Cramer’s V = 0.173) and age group (χ²=19.241, df = 2, p < 0.001; Cramer’s V = 0.280) were significantly associated with binary attitude classification. Males were more likely to be classified in the strongly positive attitudes category, and younger students (18–20 years) demonstrated the highest proportion of strongly positive attitudes compared to older age groups. Year of study (p = 0.546), previous experience with older people (p = 0.630), and intention to work with older people upon graduation (p = 0.788) were not significantly associated with attitude classification. Binary logistic regression was subsequently conducted with gender and age group entered as independent variables, with both unadjusted (COR) and adjusted odds ratios (AOR) reported. The model was statistically significant (χ²=27.992, df = 3, p < 0.001), with a Nagelkerke R² of 0.144, indicating that the model explained approximately 14.4% of the variance in attitude classification. The Hosmer-Lemeshow goodness-of-fit statistic confirmed adequate model fit (p = 0.993), and overall classification accuracy was 64.6%. Female gender was independently associated with significantly lower odds of strongly positive attitudes (AOR = 0.438; 95% CI: 0.249–0.772; p = 0.004). Compared to the youngest age group (18–20 years), both the 21–23-year group (AOR = 0.361; 95% CI: 0.201–0.648; p < 0.001) and the 24–26-year group (AOR = 0.171; 95% CI: 0.070–0.419; p < 0.001) had significantly lower adjusted odds of strongly positive attitudes. These findings are presented in Table 5. Table 5Factors associated with and predictive of positive attitudesVariableχ²df p VCOR (95% CI)AOR (95% CI)p (regression)Chi-Square AnalysisGender7.38710.007*0.173———Age Group19.2412< 0.001**0.280———Year of Study2.12830.5460.093———Previous Experience0.23210.6300.031———Intention to Work0.07210.7880.017———Binary Logistic Regression (Dependent Variable: High Positive Attitudes)Gender: Female vs. Male (ref)————0.477 (0.278–0.817)0.438 (0.249–0.772)0.004*Age: 21–23 vs. 18–20 yrs (ref)————0.375 (0.211–0.665)0.361 (0.201–0.648)< 0.001**Age: 24–26 vs. 18–20 yrs (ref)————0.185 (0.077–0.445)0.171 (0.070–0.419)< 0.001**COR = Unadjusted (Crude) Odds Ratio; AOR Adjusted Odds Ratio; V = Cramer’s V; CI = Confidence Interval. Reference categories: Gender = Male; Age = 18–20 years. Model fit: χ²=27.992, df = 3, p < 0.001; Nagelkerke R²=0.144; Hosmer-Lemeshow p = 0.993; Classification accuracy = 64.6%. *p < 0.05; **p < 0.001 Factors associated with and predictive of positive attitudes COR = Unadjusted (Crude) Odds Ratio; AOR Adjusted Odds Ratio; V = Cramer’s V; CI = Confidence Interval. Reference categories: Gender = Male; Age = 18–20 years. Model fit: χ²=27.992, df = 3, p < 0.001; Nagelkerke R²=0.144; Hosmer-Lemeshow p = 0.993; Classification accuracy = 64.6%. *p < 0.05; **p < 0.001 Discussion This study assessed the attitudes of undergraduate nursing students toward caring for older adults in Limpopo Province, South Africa. While the overall sample demonstrated a tendency toward positive attitudes, with mean scores exceeding the scale’s neutral midpoint of 119, almost all participants scored above the theoretical midpoint [27]. Yet a substantial proportion fell below the sample mean, suggesting that positive attitudes toward older adults are not uniformly held among future nurses in this context. Using the six-category classification proposed by Rathnayake, Athukorala and Siop, the overall sample demonstrated slightly positive attitudes, with the mean score falling within the slightly positive range [39]. These findings are consistent with a meta-analysis of over 30,000 nursing students globally, which found that approximately half of nursing students confirmed their willingness to participate in caring for older people [40]. These findings underscore that attitudinal ambivalence toward older adults is a persistent and cross-cultural challenge in gerontological nursing education. This observation is further supported by the finding that nursing students demonstrated a greater tendency to reject negative stereotypes than to endorse positive perceptions of older adults. If not addressed through targeted educational interventions, such attitudinal orientations may have downstream consequences for the quality and equity of geriatric care delivered by future nurses. This concern is particularly salient in the South African context, where the nursing education curriculum provides limited WIL opportunities at residential care facilities for older adults and does not systematically incorporate dedicated gerontological nursing content [16]. This structural gap limits students’ opportunities to develop and consolidate positive attitudes toward older adults across the four years of their training. The overall attitude scores obtained in the current study were comparatively higher than those reported in several international studies using the same instrument. Nursing students in Portugal, China and Hong Kong demonstrated lower mean KAOP scores than participants in the present study [41–43]. Furthermore, the mean KAOP score obtained in the current study exceeded even the highest discipline-specific score reported among health sciences students in Poland [44]. This suggests that nursing students in this study held comparatively more positive attitudes toward older adults than their counterparts across various health sciences disciplines in high-income countries. This finding is particularly noteworthy given that South Africa faces considerably greater resource constraints in nursing education than Poland, yet attitudinal outcomes were comparatively favourable. The comparatively favourable attitude scores observed in the present study may reflect contextual factors beyond formal educational preparation. One possible explanation is the influence of sociocultural values that emphasise respect, dignity, and care for older adults. In South Africa, Ubuntu promotes compassion, reciprocity, and communal responsibility toward others [45]. Similarly, studies conducted in China have suggested that filial piety may contribute to positive attitudes toward older adults among nursing students [43]. Although cultural factors were not directly measured in the present study, these findings suggest that sociocultural influences may contribute to shaping attitudes toward older adults and warrant further investigation. In the present study, gender did not emerge as a significant predictor of attitudes toward older adults in continuous analyses. However, gender was identified as a significant predictor in the binary classification model, with male students demonstrating a higher likelihood of holding strongly positive attitudes, reflecting the complexity of gender as an attitudinal variable. This finding is inconsistent with the broader literature, which predominantly reports that female gender positively influences attitudes toward older adults among nursing students [46–48]. These findings suggest that gender effects on attitudes toward older adults are likely mediated by contextual, cultural, and educational factors rather than representing a direct or universal relationship. Within the South African context, where nursing has historically been a female-dominated profession, the unexpected association between male gender and more positive attitudes in the current study may reflect unique institutional or socialisation dynamics at this particular NEI. Age group emerged as the strongest independent predictor of attitudes toward older adults in the current study. Younger students demonstrated the highest likelihood of strongly positive attitudes, which declined progressively with increasing age after controlling for gender. However, age was not significantly associated with continuous KAOP scores in the present study. In contrast, a study conducted in Portugal found that students’ age was not a significant factor influencing attitudes regarding older adults [10]. Taken together, these findings suggest that the relationship between age and attitudes toward older adults may vary across educational and sociocultural contexts and remains insufficiently understood. Consistent with the present findings, a study conducted in Hong Kong identified a significant relationship between age group and attitudes toward older adults among nursing students [42]. These findings suggest that younger nursing students may be more receptive to positive engagement with older adults during their WIL, resulting in more favourable attitudes than those observed among older students. However, the mechanisms underlying this relationship remain unclear. Further research is therefore needed to explore the factors that may explain these age-related differences in attitudes toward older adults. Notably, age effects on gerontological attitudes among nursing students in sub-Saharan Africa remain understudied, and the current finding contributes to this limited evidence base. The absence of a significant association between previous experience and attitudes is particularly noteworthy, suggesting that mere exposure without structured reflection and guided learning is insufficient to foster positive attitudinal change. This finding highlights the need for Nursing Education Institutions to incorporate structured reflective activities, guided clinical debriefing, and age-friendly learning experiences into gerontological training to maximise the educational value of interactions with older adults. This is consistent with Hammad and colleagues’ argument that purposefully designed clinical learning opportunities that incorporate age-friendly principles are essential for transforming nursing students’ attitudes toward older adults [49]. This reinforces the urgent need for intentional and structured gerontological content within the nursing curriculum. Limitations Several limitations should be considered when interpreting the findings. The cross-sectional design precludes causal inferences and does not allow for tracking of attitudinal change over time. Data were collected at a single NEI in Limpopo Province, limiting generalisability to other institutions and provinces. Self-report measurement introduces the possibility of social desirability bias, whereby students may have reported more favourable attitudes than they actually hold. The Cronbach’s alpha of 0.663 fell marginally below the conventional threshold of 0.70, although this is consistent with comparable studies in Sub-Saharan African contexts. The marginal reliability coefficient may reflect differences in the interpretation of selected KAOP items, particularly given the diverse linguistic and cultural backgrounds of participants. Furthermore, although the original English version of the KAOP scale was used without modification, varying levels of exposure to older-adult care among students may have influenced how certain items were understood and responded to. Recommendations Nursing Education Institutions should integrate dedicated gerontological nursing modules into the four-year Bachelor of Nursing curriculum. These modules should incorporate reflective learning, case-based teaching, simulation, and guided debriefing following clinical encounters with older adults to promote positive attitudes and strengthen gerontological competencies among nursing students. Accredited health facilities should ensure that WIL experiences involving older adults are structured, purposeful, and supported by facility-based gerontological mentorship programmes. Increased exposure to old age homes and other older-adult care settings may enhance students’ confidence, competence, and attitudes toward caring for older adults. The SANC should review and strengthen Regulation R174 to explicitly include gerontological nursing competencies as a core requirement for all accredited Nursing Education Institutions. Lastly, future research should prioritise longitudinal and multi-site studies, explanatory sequential mixed-methods studies exploring nursing students’ experiences of caring for older adults to provide deeper insight into the factors underlying quantitative findings, and psychometric evaluation of the KAOP Scale within South African and Sub-Saharan African contexts. Conclusion This study demonstrated that the majority of undergraduate nursing students in Limpopo Province held slightly positive attitudes toward older adults. Age group and gender were significant predictors of attitudes, with younger and male students demonstrating stronger positive orientations. Notably, year of study, previous experience, and intention to work with older adults were not significantly associated with attitudes, suggesting that academic progression and informal exposure alone are insufficient to drive meaningful attitudinal change. These findings raise concerns regarding the attainment of the objectives of the Older Persons Act 13 of 2006, which seeks to protect the rights, dignity, and social participation of older persons. When future nurses hold only slightly positive attitudes toward older adults and demonstrate no significant intention to work with this population, the responsibility of nursing education institutions to cultivate attitudes that genuinely reflect the values enshrined in the Act becomes critically urgent. The absence of dedicated gerontological nursing content within the R174 curriculum represents a structural gap that perpetuates this attitudinal deficit [32]. Systematic curriculum reform at the NEI level, supported by regulatory revision at SANC, is essential to cultivating a gerontologically competent nursing workforce capable of delivering equitable, person-centred care to South Africa’s growing older adult population. Such reform directly advances SDG 3, which calls for ensuring healthy lives and promoting well-being for all ages, and SDG 4, which emphasises inclusive and quality education, while aligning with Africa Agenda 2063’s vision of a healthy and well-nourished continent in which no older person is left behind. Acknowledgements The authors gratefully acknowledge the undergraduate nursing students who participated in this study for their time and willingness to share their perspectives. The authors also acknowledge Chokoe MP, Senoamadi TE, Sekgobela PA, Pholoana PK, and Monkwe KJ, final-year undergraduate nursing students from the Class of 2024, who undertook the original research project under the supervision of TAN, from which this manuscript was developed. Clinical trial number Not applicable. Abbreviations Nursing Education Institution :Nursing Education Institutions South African Nursing Council Sustainable Development Goal Turfloop Research Ethics Committee Work-Integrated Learning Authors’ contributions TAN was responsible for the research design and conceptualisation, conducted the study, including data collection and fieldwork supervision, and led the writing and editing of the manuscript. PMM performed the statistical analysis and interpretation of data and contributed to the writing and editing of the manuscript. LN and TAP were invited as co-authors and contributed to the critical review, writing, and editing of the manuscript. All authors read and approved the final manuscript and take responsibility for the final content. Funding This study received no external funding. Data availability The datasets generated and analysed during the current study are not publicly available due to participant confidentiality requirements, but are available from the corresponding author on reasonable request, subject to approval by the TREC of the University of Limpopo. Declarations Ethics approval and consent to participate Ethical approval for this study was granted by the TREC of the University of Limpopo, South Africa (Reference: TREC 1447/2024:UG). The study was conducted in accordance with the Declaration of Helsinki. Informed consent was obtained from all participants before data collection, and participation was voluntary and anonymous. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The datasets generated and analysed during the current study are not publicly available due to participant confidentiality requirements, but are available from the corresponding author on reasonable request, subject to approval by the TREC of the University of Limpopo.