Do Educators’ Demographic Characteristics Drive Learner Academic Performance? Examining the Role of Gender, Qualifications, and Experience

Abstract

The influence of educators’ demographic characteristics on learner academic performance has garnered attention, particularly in developing countries. However, gaps remain in understanding the relationship between demographic factors, professional qualifications, and student academic performance, especially in under-resourced educational environments; this study examines the impact of educator demographics and qualifications on learner performance. The study investigates four key objectives: (i) analysing the relationship between educator qualifications and learner academic performance, (ii) evaluating the effect of educator demographic factors on learner academic performance, (iii) comparing the relative influence of demographic characteristics and professional qualifications, and (iv) assessing the validity of Mincer’s Earnings Function in capturing diminishing returns to experience. Using secondary data from the South African School Administration and Management System (SA-SAMS), this study examines 70 educators across five secondary schools. An Ordinary Least Squares (OLS) regression model is employed for estimation, with Robust Least Squares (RLS) used as a validation technique to ensure result consistency. The empirical results reveal that teaching experience and gender significantly influence learner performance, whereas age and formal qualifications exhibit no statistical effect. The findings further confirm diminishing returns to experience, indicating that while teaching experience enhances academic performance, its marginal impact declines beyond a certain threshold.

1. Introduction

The transformation of the South African educational system marks a shift from past practices, aiming to enhance the role of educator input factors in improving learner academic performance (Nnadozie, 2024; Ncanywa et al., 2022). The reformation intends to meet learners’ needs and ensure their potential is fully realised, reducing societal inequalities (Ngobeni et al., 2023). In addition, the quality of education remains one of the primary drivers of socioeconomic development, particularly in developing nations like South Africa, where inequalities in academic performance continue to show structural differences (Ogujiuba et al., 2024; Xholo et al., 2025). However, the empirical literature identifies numerous factors influencing student success, ranging from teacher quality, school resources, parental involvement, household income, peer influences, school culture, and government policies (Tan, 2024; Careemdeen, 2024). Nevertheless, educators remain one of the most important determinants of academic performance (Sok & Heng, 2024). Educators are fundamental to student academic success, as they deliver curriculum content, promote critical thinking, and create engaging learning environments. More so, effective teachers can lessen socioeconomic disadvantages through innovative and inclusive practices while serving as mentors who help to develop students’ motivation and aspirations.

Theoretically, the relationship between educator characteristics and learner academic performance is rooted in various educational and human capital theories. The Human Capital Theory (Becker, 1964) posits that investments in teacher qualifications and professional development enhance their effectiveness, improving student learning performance. Likewise, this aligns with the argument that well-trained educators possess superior pedagogical skills, content mastery, and adaptive teaching strategies, directly influencing academic achievement (Baikady, 2025; Edgar, 2012). Additionally, the Teacher Effectiveness Framework stresses the role of instructional quality, suggesting that professional training contributes more to student success than inherent demographic traits such as age or gender (Kim et al., 2018; Costantine et al., 2025). The Expectancy–Value Theory (Eccles & Wigfield, 2023) further supports this connection by emphasising that educators with higher qualifications and experience are more likely to promote student motivation, engagement, and performance through structured learning environments and mentorship. Conversely, the empirical literature remains inconclusive regarding the influence of educator demographics (Chung et al., 2022; Redding, 2019; Gentrup et al., 2020; Lowther et al., 2003; Redding & Carlo, 2025), with some studies suggesting that age and gender may influence teaching styles and student-teacher interactions (Roorda & Jak, 2024; Mullola et al., 2011), while others find no significant impact (Francisco, 2020; Leech & Haug, 2015).

The exploration of this study is justified in four key ways: (i) it addresses persistent educational inequalities in South Africa, particularly under-resourced regions like Amathole East; (ii) it clarifies the role of educator demographics and qualifications, distinguishing whether professional training impacts student performance rather than attributes like age, gender, and experience; (iii) it informs policy on teacher recruitment, training, and professional development; and finally, (iv) it contributes to the empirical literature and aligns with global educational goals.

First, South Africa’s education system continues to be faced with inequalities, particularly in under-resourced regions like Amathole East, where inequalities in teacher quality, school infrastructure, and learning remain a pressing challenge (Bayat et al., 2024; Chirowamhangu, 2024). Socioeconomic factors, including poverty, limited access to educational resources, and inefficiencies, contribute to differences in student performance across the country, making educators’ role more important in addressing the situation. A well-trained and highly qualified teacher can help reduce these disadvantages by promoting an inclusive and effective learning environment. However, policy efforts to improve teaching quality risk being misdirected without empirical clarity on which educator attributes have the most impact on student achievement. Therefore, this study aims to add to the empirical literature on how this can be addressed by examining how educator characteristics influence academic performance in one of South Africa’s most vulnerable regions. This study provides insights to support policies by identifying whether teacher qualifications and demographic factors contribute to student success. Moreover, improving the quality of education in regions like Amathole East is crucial for encouraging long-term social mobility, reducing educational inequities, and ensuring that all students, regardless of their socioeconomic background, can excel academically.

Secondly, another essential aspect of educational studies is the factors that influence student achievement, particularly the extent to which educator characteristics influence learning performance (Brink et al., 2021). While existing studies emphasise the importance of teacher quality (Snoek, 2021; Cochran-Smith, 2021; Witter & Hattie, 2024; Murtonen et al., 2024), there remains a lack of clarity regarding the relative impact of professional qualifications versus demographic attributes such as age, gender, and years of experience. A distinction regarding this is essential, as policies that improve educational performance must be guided to enhance teaching effectiveness. In this view, this study provides much-needed insight by focusing on this area. For example, suppose qualifications prove to be the key determinant. In that case, it shows the necessity of investing in teacher education and continuous professional development rather than relying on tenure or demographic diversity as primary effectiveness indicators. Conversely, if demographic factors play a more critical role, it may suggest that experience-driven pedagogical strategies or gender-sensitive teaching approaches require greater emphasis in teacher preparation programmes.

Thirdly, effective education policies focus on the factors driving student success (Timotheou et al., 2023). Given the prevalence problem in South Africa’s education sector, low academic performance in South Africa is believed to be connected with socioeconomic challenges (Spaull, 2013; Woldegiorgis & Chiramba, 2024). The Eastern Cape, where the Amathole East District is situated, demonstrates this problem with its consistently low pass rates in national exams (Kohli, 2023). In 2019, South Africa’s National Senior Certificate (NSC) examination achieved a pass rate of 81.3%; this figure declined to 76.2% in 2020, marking a 5.1 percentage point decrease attributed to the challenges posed by the COVID-19 pandemic. In 2021, the pass rate experienced a slight increase, reaching 76.4% (Department of Basic Education, 2025). The Amathole East District’s educational performance has exhibited fluctuations in recent years. In 2020, the district recorded a pass rate of 67.3%, which increased to 74.4% in 2021. However, in 2022, the pass rate experienced a decline, dropping to 70.4% (Province of Eastern Cape Education, 2023). Compared to other districts within the Eastern Cape province, Amathole East’s performance has been unstable. For instance, in 2022, the leading district in the province was Alfred Nzo East, which achieved a pass rate of 77.0%; this indicates that while Amathole East has made progress in specific years, there remains a performance gap when compared to the top-performing districts in the region (Province of Eastern Cape Education, 2023).

Regarding this performance, effective education policies are needed; one of the most important aspects of education reform is ensuring that teacher recruitment, training, and professional development align with what makes educators effective (Smith & Gillespie, 2023; Patfield et al., 2023; Zhang et al., 2024). This study tends to contribute to the guidance by identifying whether professional qualifications or demographic characteristics have a greater influence on student performance. It contributes to existing studies to inform policymaking efforts to strengthen the teaching staff and enhance learning performance. Additionally, this study points to the need for continuous professional development to sustain teacher effectiveness. Education systems must move beyond initial teacher certification and invest in lifelong learning opportunities that allow educators to improve their methodologies, integrate innovative instructional techniques, and address educational challenges; this study stressed that recruitment, training, and professional development policies are data-driven and must be aligned with long-term educational goals to improve student success.

Finally, this study contributes to the empirical literature by filling the identified gap. Existing research has examined various determinants of student success. For example, studies investigate the determinants of students’ success at university (Hattie, 2023; Danilowicz-Gösele et al., 2017; Naylor & Smith, 2004; Akimov et al., 2024). Other studies have looked at college retention and family and community (Beard & Thomson, 2021), but there remains a gap in understanding how demographic attributes and professional qualifications influence learning performance regarding educational inequality. Focusing on South Africa’s Amathole East District, this study analyses whether educator qualifications hold greater predictive power for academic achievement than demographic factors such as age, gender, and teaching experience. The findings enrich the debate on teacher effectiveness and inform discussions on equitable access to quality education in developing economies. Beyond its empirical contributions, this study aligns with global educational priorities, particularly the United Nations Sustainable Development Goal (SDG) 4, which emphasises inclusive and equitable quality education for all (Asaleye & Ncanywa, 2025). Well-trained and qualified teaching staff is vital for addressing disparities in students’ learning and ensuring that students, regardless of socioeconomic background, receive the support necessary for academic success (Rodriguez et al., 2025). The study pointed out the importance of professional qualifications over other demographic characteristics, which is helping to meet the need for policies that prioritise teacher recruitment, training, and continuous professional development. The national education strategies may benefit by improving learning performance, reducing disparities, and advancing global commitment to quality education.

From the foregoing discussion, the main objective of this study is to examine the influence of educator demographic characteristics (gender, age, and experience) and professional qualifications on learner academic performance in an under-resourced educational environment. Specifically, the study seeks to accomplish the following:

i.

Analyse the relationship between educator qualifications and learner academic performance.

ii.

Evaluate the effect of educator demographic factors (gender, age, and experience) on learner academic performance.

iii.

Compare the relative influence of demographic characteristics and professional qualifications on learner academic performance.

iv.

Assess the validity of Mincer’s Earnings Function in capturing diminishing returns to experience.

The structure of this study is organised as follows: Following the Introduction, Section 2 outlines the materials and methods. Section 3 presents the results, and Section 4 discusses the results and evaluates the hypotheses. Finally, Section 5 concludes the study with a summary of the work and policy recommendations.

2. Materials and Methods

2.1. Theoretical Framework and Model Specification

2.1.1. Theoretical Framework

This study is grounded in the Human Capital Theory (HCT) (Becker, 1964), which posits that investment in education and training enhances individual productivity and performance. Relating this theory to this study, the quality of teachers measured through their qualifications, experience, and training is expected to influence learner academic performance directly. The fundamental principle of the HCT can be represented as follows:

$$Y=f(H)$$

(1)

In Equation (1), $Y$ indicates an individual’s productivity (in this case, learner academic performance) and $H$ represents the accumulated human capital (educator qualifications, experience, and training). The underlying assumption is that higher teacher qualifications lead to improved instructional delivery, enhanced student engagement, and better academic performance. The equation is extended further by incorporating Mincer’s Earnings Function (Mincer, 1974), which stresses experience as an essential factor influencing productivity. Therefore, we have

$$Y={\alpha }_0+{\alpha }_{1}S+{\alpha }_{2}EXP+{\alpha }_{3}EX{P}^2+e_t$$

(2)

In Equation (2), $S$ represents years of schooling (qualification level), $EXP$ denotes work experience, and $EX{P}^2$ accounts for diminishing returns to experience. In an educational environment, this suggests that teaching experience contributes to effectiveness, but its impact may reduce over time.

Equation (2) is modified with the inclusion of the Teacher Effectiveness Framework by Stronge (2018), which suggests that teacher demographics, including age and gender, influence instructional effectiveness. While younger teachers may exhibit greater adaptability to modern pedagogical methods, more experienced teachers may possess more substantial content mastery. Additionally, gender-based differences in teaching styles may impact student engagement and performance (Jamil & Raza, 2024; Nkepah, 2025). The impact of these variables on learner academic performance is given as follows:

$$LAB=f(Qualification,Experience,Age,Gender)$$

(3)

In Equation (3), the independent variables (qualification, experience, age, and gender) collectively determine learner performance.

2.1.2. Model Specification

Based on the theoretical framework, the relationship between learner academic performance (LAB) and educator characteristics can be expressed functionally as follows:

$$LAB=f(AGE,GENDER,QUAL,EXP)$$

(4)

In Equation (4), all other variables hold the same definition, and $QUAL$ and $EXP$ are qualifications and experience, respectively. Gender is estimated using male = 1 and female = 0. Equation (4) is transformed and written explicitly as follows:

$$LA{B}_t={\beta }_0+{\beta }_{1}AGE+{\beta }_{2}GENDER+{\beta }_{3}QUAL+{\beta }_{4}EXP+{\beta }_{5}EX{P}^2+{\epsilon }_t$$

(5)

The expected signs of the coefficients in Equation (5) are as follows: ${\beta }_3>0$: this is due to higher qualifications, which should improve academic performance, supporting the Human Capital Theory; ${\beta }_4>0$: this is a result of experience, which is expected to positively impact student performance, following Mincer’s framework; and ${\beta }_5<0$: this is because of diminishing returns to experience. However, for ${\beta }_1$ and ${\beta }_2$, that is, the impact of age and gender, this is theoretically ambiguous, requiring empirical validation. ${\epsilon }_t$ is the error term and ${\beta }_0$ is the intercept or constant term.

2.2. Development of Hypotheses and Techniques of Estimation

2.2.1. Development of Hypotheses

The empirical literature has reported the role of educators in influencing learner academic performance, with both professional qualifications and demographic characteristics serving as potential determinants (Mosia et al., 2025; Dahri et al., 2024; Tadese et al., 2022); this study formulates its hypotheses based on the Human Capital Theory (Becker, 1964) and the Teacher Effectiveness Framework (Stronge, 2018).

First, educator qualifications are often regarded as a fundamental determinant of learner achievement. Higher academic and professional credentials enhance pedagogical competence, subject-matter expertise, and instructional delivery, improving student engagement and performance. Drawing on the Human Capital Theory, which posits that education and training enhance worker productivity (Becker, 1964), this study examines how educator qualifications and demographic attributes affect learner academic performance. The Human Capital Theory suggests that formal training improves teaching quality and improves student outcomes. Accordingly, the following hypotheses are formulated:

H₁: 

There is a statistically significant relationship between educator qualifications and learner academic performance.

In addition to formal qualifications, several demographic variables such as gender, age, and teaching experience have been associated with variations in instructional effectiveness. For instance, gender may influence teaching style and classroom interaction (Stolk et al., 2021), while age may be associated with receptivity to modern pedagogical methods (Bai & Zang, 2025). Moreover, Mincer’s (1974) Earnings Function shows the value of work experience in enhancing productivity, though marginal returns may decrease over time. The following hypotheses are therefore proposed:

H₂: 

There is a statistically significant relationship between educator gender and learner academic performance.

H₃: 

There is a statistically significant relationship between educator age and learner academic performance.

H₄: 

There is a statistically significant relationship between educator experience and learner academic performance, with a diminishing effect at higher levels of experience.

The existing literature presents mixed findings regarding the relative influence of educator qualifications versus demographic characteristics. Some scholars emphasise the primacy of professional credentials in determining teaching effectiveness (Sims & Fletcher-Wood, 2021; Tatto, 2021), while others reported the role of individual characteristics and classroom experience (Even & BenDavid-Hadar, 2025; Stevanović et al., 2021); this study aims to contribute to this discourse by empirically assessing the comparative influence of both sets of variables.

H₅: 

Educator qualifications explain a greater proportion of the variance in learner academic performance than educator demographic characteristics (age, gender, and experience).

2.2.2. Techniques of Estimation and Information About the Data

The study achieved its objectives by employing Ordinary Least Squares (OLS) regression and Robust Least Squares (RLS) analysis. Inferential statistics were used to conclude the sample data. Specifically, a correlation analysis examines the strength and direction of the relationships between key demographic characteristics (age, gender, experience, qualifications, and learner academic performance); this preliminary step helps identify patterns and potential associations before regression analysis. We adopted the OLS regression technique as the primary estimation method to determine the statistical significance of these relationships. The model specification follows a linear functional form, where learner academic performance is expressed as a function of educator age, gender, experience, and qualifications. OLS is particularly suitable for this study as it minimises the sum of squared residuals, ensuring an optimal linear fit. The regression results are further subjected to diagnostic checks, including the normality of residuals, ensuring the reliability and validity of the estimates. The OLS model is given as follows:

$$Y_t={\beta }_0+{\displaystyle \sum _{i=1}^n{\beta }_1{X}_{it}}+{\epsilon }_t$$

(6)

In Equation (6), $Y_t$ represents the dependent variable, that is, learner academic performance; $X_{it}$ denotes the explanatory variables (age, experience, qualification, and gender); ${\beta }_0$ is the intercept; ${\beta }_1$ denotes the estimated coefficients of the independent variables; and ${\epsilon }_t$ is the error term. Likewise, to account for potential heteroscedasticity and the influence of outliers, we estimate the model using Robust Least Squares (RLS) as a robustness check, which minimises the impact of extreme observations. The robust estimation follows the iteratively reweighted least squares (IRLS) approach as given:

$$Y_t={\alpha }_0+{\displaystyle \sum _{i=1}^n{\alpha }_1{X}_{it}}+u_t$$

(7)

In Equation (7), ${\alpha }_0$ is the intercept, ${\alpha }_1$ denotes the estimated coefficients of the independent variables, and ${\epsilon }_t$ is the error term. $u_t$ represents a disturbance term adjusted by a weighting function $w_t$, which reduces sensitivity to outliers. The estimation is performed using Huber’s M-estimator, which assigns weights as follows:

$$w_t=\left\{\begin{array}{ll}1,& if\left|{\epsilon }_t\right|\le c\sigma \\ {\displaystyle \frac{c\sigma }{\left|{\epsilon }_t\right|}}& if\left|{\epsilon }_t\right|>c\sigma \end{array}\right.$$

(8)

In Equation (8), ‘c is a tuning constant (typically 1.345 for Huber’s function), and σ is an estimate of the standard deviation of residuals. The key model assumptions were tested to ensure the regression analysis’s validity. For example, the independence of observations was verified through dataset design and the Durbin–Watson statistics. Likewise, the normality of residuals was evaluated through histograms (Asaleye et al., 2023). The result of the histogram normality test is presented in Appendix A.

The data for this study were sourced from the South African School Administration and Management System (SA-SAMS) database. This government-managed repository consolidates extensive demographic and academic performance records of educators and learners across South Africa. This dataset includes comprehensive information on 70 educators from five regional secondary schools in the Amathole East District. The South African School Administration and Management System (SA-SAMS) is this study’s primary data source. It provides detailed, systematically curated data on key variables, including educator age, gender, qualifications, years of experience, and learner academic performance, typically measured through standardised assessment scores (DBE, 2025). A purposive sampling strategy was employed to select five secondary schools, ensuring the inclusion of diverse education within the district. While this approach supports contextual relevance, the study acknowledges the limitations inherent in a small sample size. SA-SAMS was selected due to its comprehensiveness and capacity to support robust statistical analyses and evidence-based evaluations of factors influencing learner performance (DBE, 2025).

The validity and reliability of the data used in this study are ensured through its source, the South African School Administration and Management System (SA-SAMS). Education authorities manage this system and follow standardised data collection procedures, ensuring consistency and accuracy. Furthermore, to enhance the robustness of the analysis, we employed both Ordinary Least Squares (OLS) and Robust Least Squares estimation techniques. These methods account for potential heteroskedasticity and outliers, improving the reliability of the results.

3. Presentation of Results

3.1. Preliminary Analysis

3.1.1. Frequency of Demographic Data

Table A1. Demographic characteristics of study participants.

Demographic Variables	Category	Number (N)	Percentage
Gender	Male	15	21%
	Female	55	79%
Age of Educators	20 to 30	5	7%
	31 to 40	7	10%
	41 to 50	31	44%
	51 to 60	27	39%
Work Experience	1 to 10	17	24%
	11 to 20	33	47%
	21 to 30	12	17%
	31 upwards	8	12%
Educational Qualification	Professional Teaching	23	33%
	Post-Professional Degree	44	63%
	PGCE	3	4%

in Appendix A shows the demographic characteristics of study participants used in this study. In Figure 1, the demographic analysis reveals that 44% of participants were aged 41–50, making it the largest group; 39% were older (51–60 years), and 17% were younger (7% aged 20–30; 10% aged 31–40). The diverse age distribution combines seasoned educators with newer entrants, enriching teaching perspectives. Experienced educators may employ effective strategies and manage the classroom better, while younger teachers introduce fresh ideas; this blend of age demographics is crucial for enhancing academic performance in secondary schools, particularly in the Amathole East District (Leech & Haug, 2015). Studies show that younger educators offer innovative methods, whereas seasoned teachers have superior classroom management skills and subject knowledge, positively impacting student performance (Lucksnat et al., 2024; Martin-Alguacil et al., 2024).

Figure 2 shows the gender distribution of educators in the sample. The demographic analysis reveals 79% (n = 55) female and 21% (n = 15) male respondents. The studies by Mulawarman and Komariyah (2021) and Priadi et al. (2023) state that female educators typically use collaborative and nurturing teaching styles, while males may be more assertive. However, the effectiveness of these approaches can vary based on student needs (Abou-Khalil et al., 2021).

Figure 3 indicates the work experience of educators. The demographic analysis indicates varied work experience among respondents in the study. Most participants had 11–20 years of experience, followed by 1–10 years, 21–30 years, and 31 and more; this distribution shows a range of experience levels among educators in the Amathole East District.

Figure 4 demonstrates the education qualification in the sample. The demographic analysis of the study reveals that 63% (n = 44) of participants possess professional teaching qualifications (in South Africa, possessing professional teaching qualifications refers to holding a recognised credential, such as a Bachelor of Education (BEd) or a Postgraduate Certificate in Education (PGCE), which certifies an individual’s competence to teach in formal educational schools), indicating a well-qualified respondent group focused on effective instructional practices. Additionally, 33% (n = 23) hold other educational qualifications, including post-professional degrees like B Ed Honours and Master of Education, showcasing a commitment to further education and professional development. The remaining 4% (n = 3) have a PGCE, reflecting another pathway into the teaching profession.

3.1.2. Descriptive Statistics and Correlation Analysis

Table 1. Summary of the descriptive statistics.

	Observation	Mean	Std. Error	Std. Deviation	Variance	Skewness	Kurtosis
Age	70	3.14	0.104	0.873	0.762	0.150	−1.000
Experience	70	2.16	0.111	0.927	0.859	0.250	−0.800
Qualification	70	1.41	0.069	0.577	0.333	0.002	−1.200
Gender	70	1.21	0.049	0.413	0.171	1.400	0.001
Academic Performance	70	2.53	0.139	1.164	1.354	−0.350	−0.500

Source: authors’ computation.

presents the descriptive statistics for the 70 participants selected per school in the Amathole East District, which provides insights into the central tendency and dispersion of key demographic variables. The mean age of respondents is 3.14 with a standard deviation of 0.873, indicating that while the average age falls within a specific range, there is moderate variability among participants. Work experience has a mean of 2.16 and a standard deviation of 0.927, suggesting that respondents, on average, have a relatively low level of experience, with some variation across the sample. Regarding educational qualifications, the mean value of 1.41 and a standard deviation of 0.577 imply that most respondents possess a basic-to-intermediate level of formal education, with minimal dispersion from the mean. The gender variable has a mean of 1.21 and a standard deviation of 0.413, reflecting a distribution skewed toward a particular gender category while maintaining a relatively low level of variability. Lastly, academic performance has a mean of 2.53 and a standard deviation of 1.164.

The Pearson correlation analysis in

Table 2. Correlation matrix.

	Academic Performance	Gender	Age	Experience	Qualification
Academic Performance	1000
Gender	0.002 (0.9868)	1000
Age	0.124 (0.3664)	0.155 (0.2001)	1000
Experience	−0.051 (0.6750)	0.062 (0.6101)	0.258 (0.0311)	1000
Qualification	0.317 (0.0075)	−0.135 (0.2652)	0.025 (0.8372)	−0.015 (0.9019)	1000

The probability values are in the brackets. Source: authors’ computation.

provides an initial exploration of linear relationships among the variables in this study. Notably, only the relationships between educator qualifications and learner academic performance, as well as between age and teaching experience, are statistically significant. As the primary focus here is descriptive, the correlation matrix serves as a preliminary diagnostic tool, and not as a basis for inferential conclusions. The correlation coefficient of 0.002 for gender indicates an almost negligible association with learner academic performance, suggesting that gender plays no substantial role in influencing educational performance. Similarly, the correlation of 0.124 between age and learner performance reflects a very weak positive relationship, while the coefficient of −0.051 for experience indicates a very weak negative association. These low values suggest that age and experience, considered independently, contribute minimally to the variation in learner academic performance.

In contrast, the correlation of 0.317 between educator qualifications and learner academic performance reflects a moderate positive relationship, implying that higher teacher qualifications are moderately associated with improved academic outcomes. This aligns with the work of Harbatkin et al. (2025), who found that teacher qualifications are a key determinant of student achievement. Their study shows how advanced qualifications often enhance pedagogical expertise, deepen content mastery, and promote more effective instructional strategies, all contributing to learner success.

3.2. Regression Analysis

Table 3. Least squares and robust least squares estimates.

Least Squares Estimate
Included Observations: 70
Dependent Variable: Learner Academic Performance
Variable	Coefficient	Std. Error	t-Statistic	Prob.
AGE	0.193841	0.123898	1.564516	0.1226
EXPERIENCE SQUARED	−0.14207	0.054054	−2.628285	0.0107
EXPERIENCE	0.196921	0.104178	1.890246	0.0633
QUALIFICATION	0.183375	0.445577	0.411545	0.682
MALE = 0	1.15718	0.522863	2.213164	0.0305
MALE = 1	1.16328	0.525677	2.212915	0.0305
R-squared	0.51497	Durbin–Watson stat		2.199801
Adjusted R-squared	0.483367
Normality Test	Jarque–Bera	2.822708	Prob.	0.243813
Robust Least Squares Estimate
Included Observations: 70
Dependent Variable: Learner Academic Performance
Variable	Coefficient	Std. Error	z-Statistic	Prob.
AGE	0.193731	0.127379	1.520906	0.1283
EXPERIENCE SQUARED	−0.136903	0.055573	−2.463492	0.0138
EXPERIENCE	0.184297	0.107104	1.720726	0.0853
QUALIFICATION	0.240479	0.458094	0.524955	0.5996
MALE = 0	1.096236	0.53755	2.039318	0.0414
MALE = 1	1.103899	0.540444	2.042577	0.0411
R-squared	5.126931	Rw-squared	5.16527
Adjusted R-squared	4.058723	Prob.	0.0000
Normality Test	Jarque–Bera	2.89868	Prob.	0.234725

Source: authors’ computation.

presents the Ordinary Least Squares (OLS) estimation results for the learner academic performance equation. The findings indicate that experience, experience squared, and gender are statistically significant predictors, while age and qualification exhibit no significant effect. Specifically, the coefficient for experience squared is −0.14207 (p = 0.0107), confirming diminishing returns to experience. The coefficient for experience is 0.196921 (p = 0.0633), suggesting a positive, albeit marginally significant, impact on learner performance. Gender differences are evident, with females (coded as 0) representing a coefficient of 1.15618 (p = 0.0305) and males (coded as 1) yielding 1.16328 (p = 0.0305), indicating a statistically significant effect on performance. The model demonstrates a reasonable fit, with an R-squared of 0.51497 and an adjusted R-squared of 0.483367, suggesting that the model explains approximately 48.3% of the variation in learner academic performance. The Durbin–Watson statistics (2.199801) indicate no autocorrelation concerns, while the Jarque–Bera normality test (2.822708, p = 0.243813) confirms that the residuals follow a normal distribution. The Jarque–Bera normality result is presented in Figure A1 in Appendix A.

To ensure the reliability of the OLS estimates, a robustness check was conducted using Robust Least Squares (RLS), with results also presented in Table 3. The findings confirm that experience, experience squared, and gender remain statistically significant, while age and qualification do not exhibit a significant effect. Specifically, experience squared has a coefficient of 0.184297 (p = 0.0138), strengthening the presence of diminishing returns to experience. The coefficient for experience is 0.196921 (p = 0.0853), maintaining its positive but marginal significance. Gender effects persist, with females (coded as 0) showing a coefficient of 1.096236 (p = 0.0414) and males (coded as 1) at 1.103899 (p = 0.0411), suggesting a statistically significant influence on learner performance.

The model exhibits explanatory power, with an R-squared of 5.126931 and an adjusted R-squared of 4.058723. The Jarque–Bera normality test (Jarque–Bera = 2.89868, p = 0.234725) confirms the normal distribution of residuals; the result is presented in Figure A2 in Appendix A. Additionally, the robust Wald test (Rw-squared = 5.16527, p = 0.0000) supports the overall significance of the model; this shows the robustness of the estimation and validates the use of RLS in addressing potential heteroscedasticity and outlier effects.

The variance inflation factor (VIF) analysis in

Table A2. Variance inflation factors.

Variance Inflation Factors
Included Observations: 70
	Coefficient	Uncentered	Centred
Variable	Variance	VIF	VIF
AGE	0.015351	941.0617	2.685725
EXPERIENCE SQUARED	0.002922	139.9551	4.79229
EXPERIENCE	0.010853	319.307	2.49493
MALE = 1	0.000296	1.30058	1.040464
MALE = 0	0.000678	5.202321	0.020103
QUALIFICATION	0.198539	5782.749	1.376684
C	0.273385	6004.058	NA

in Appendix A reveals that multicollinearity is not a significant concern in the model. The centred VIF values for age (2.69), experience (2.49), experience squared (4.79), gender (1.04 for male = 1; 0.02 for female = 0), and qualification (1.38) remain below the conventional threshold of 5, indicating the absence of severe collinearity. The slightly elevated VIF for experience squared (4.79) suggests a moderate correlation with experience, which is expected due to the polynomial structure but does not pose a critical multicollinearity issue. The low VIF values for the binary variable male (1.04 for male = 1 and 0.02 for female = 0) confirm that gender does not strongly correlate with other regressors, ensuring its independent contribution to the model.

3.3. Ethical Considerations

Ethical approval for this study was obtained from the institutional research ethics committee before data collection (Approval No: FEDSECC021-03-2023). The study used secondary data from the South African School Administration and Management System (SA-SAMS). This government-managed database provides de-identified and systematically curated information on educators and learners. Access to this dataset was granted by the Department of Basic Education (DBE), and all data were handled in compliance with national regulations and institutional ethical standards. As the dataset contains anonymised information with no personally identifiable details and no direct interaction with participants, individual informed consent was not required.

4. Discussion of Results and Evaluation of Hypotheses

Hypothesis One:

Educator qualifications significantly positively impact learner academic performance.

The hypothesis cannot be accepted, suggesting that educator qualifications alone are not a sufficient predictor of learner academic performance; this stresses the importance of experience, teaching quality, and institutional support in influencing student performance.

The estimation results of the Ordinary Least Squares (OLS) and Robust Least Squares (RLS) presented in Table 3 indicate that educator qualifications do not exhibit a statistically significant effect on learner academic performance. The non-significance of qualifications suggests that, within the studied sample, differences in formal educational attainment among educators do not translate into measurable variations in student achievement. This finding challenges the conventional expectation that higher qualifications directly enhance instructional effectiveness (Mosia et al., 2025). From a theoretical perspective, the Human Capital Theory posits that increased educator qualifications should enhance pedagogical skills and instructional delivery, thereby improving student performance (Becker, 1964). However, the empirical results suggest that other factors, such as teaching experience, pedagogical strategies, or institutional resources, may influence learner performance more than formal qualifications alone (Dahri et al., 2024). The absence of a significant effect may also indicate that the quality and relevance of qualifications, rather than merely possessing higher credentials, are more critical in determining teaching effectiveness (Stronge, 2018; Scott et al., 2024).

The findings have implications for educational policy and teacher development. If qualifications alone do not significantly impact student performance, policymakers should reconsider relying solely on qualification-based salary structures or promotions. Instead, professional development programmes focusing on practical teaching methods, subject mastery, and student engagement techniques may yield greater improvements in learning performance (Stronge, 2018). Likewise, the non-significant effect of qualifications stresses the need for ongoing teacher training and pedagogical adaptation. Rather than assuming that formal education credentials suffice, education systems should emphasise workshops, mentorship programmes, and in-service training to enhance instructional efficacy (Tadese et al., 2022). Moreover, the findings suggest that factors beyond individual qualifications, such as classroom resources, student demographics, and school infrastructure, may be more prominent in influencing learner academic performance (Mosia et al., 2025; Tadese et al., 2022).

The rejection of Hypothesis One suggests that educator qualifications alone are not a sufficient predictor of learner academic performance; this stresses the importance of experience, teaching quality, and institutional support in influencing student performance. Educational stakeholders should adopt an approach to teacher effectiveness, balancing qualification requirements with practical training and pedagogical support.

Hypothesis Two:

Educator gender has a significant effect on learner academic performance.

The results confirm Hypothesis Two, indicating that educator gender statistically affects learner academic performance. However, rather than suggesting an inherent advantage of one gender over the other, the findings emphasise the importance of gender diversity in teaching. Educational policies should promote a balanced teaching staff and control gender-based strengths to enhance pedagogical effectiveness and student success.

The OLS and RLS estimation results in Table 3 indicate that educator gender is statistically significant in explaining variations in learner academic performance. Specifically, both male and female educator coefficients are statistically significant at conventional levels (p < 0.05), suggesting that educator gender plays a role in influencing student achievement. The positive and significant coefficient values imply that gender-based differences in teaching styles, mentorship approaches, or interaction patterns may influence student learning performance (Stolk et al., 2021). Theoretically, gender-based pedagogical differences have been widely discussed in educational research. Studies suggest that female educators often adopt nurturing and student-centred teaching methods, which may promote a more inclusive learning environment. At the same time, male educators may emphasise discipline and structured learning techniques that can enhance academic performance (Bai & Zang, 2025).

The statistical significance of both gender categories in this study suggests that different gendered teaching approaches may contribute uniquely to student performance rather than one gender being inherently superior. The significance of educator gender in learner academic performance has several important implications for teacher recruitment, training, and policy formulation. The findings support the need for a diverse teaching staff where both male and female educators contribute complementary skills and instructional strategies (Stolk et al., 2021). Policymakers should promote gender diversity in teacher recruitment to ensure students benefit from varied teaching approaches. Additionally, given that male and female educators significantly impact student performance, teacher training programmes should acknowledge gender-based teaching styles and develop strategies that influence these differences for improved learning performance (Bai & Zang, 2025). Encouraging collaborative teaching models may help integrate the strengths of different gender perspectives, ultimately fostering a more effective learning environment.

Hypothesis Three:

Educator age has a significant effect on learner academic performance.

The findings do not support Hypothesis Three, as educator age is not a statistically significant determinant of learner academic performance; this suggests that teaching effectiveness is more dependent on experience, qualifications, and pedagogical skills than age itself.

The OLS and RLS results presented in Table 3 indicate that educator age is not statistically significant in explaining variations in learner academic performance. The estimated coefficient for age fails to reach conventional significance levels (p > 0.05), suggesting that age alone does not have a direct and measurable impact on student achievement; this finding challenges the assumption that older educators, by their maturity or years of experience, inherently contribute more effectively to student learning performance (Stolk et al., 2021). The lack of statistical significance implies that age, as an isolated factor, may not be a reliable determinant of teaching effectiveness; this result aligns with studies suggesting that effective teaching is more closely linked to professional experience, pedagogical skills, and qualifications rather than chronological age (Bai & Zang, 2025). While older educators may bring greater classroom management skills and deeper subject knowledge, younger educators may introduce innovative teaching methods and a greater adaptability to modern educational technologies. One possible explanation for this finding is that other factors, such as professional training, adaptability to curriculum changes, and teaching methodologies, may offset age-related advantages in teaching. For instance, an older educator with limited professional development may be less effective than a younger educator who actively engages in continuous learning and instructional innovation (Mincer, 1974; Langer, 2025).

The results suggest that educational policies should not assume that older teachers are inherently better instructors. Instead, professional development programmes should focus on teaching competencies, instructional methods, and student engagement strategies to enhance learning performance across all age groups. Likewise, since age does not independently determine effectiveness, teacher recruitment strategies should emphasise qualifications, experience, and professional training rather than age as a selection criterion; this approach ensures that schools hire competent educators who can positively influence student achievement, regardless of their age (Bai & Zang, 2025). In addition, since both younger and older educators can contribute uniquely to student learning, policymakers should encourage a culture of lifelong learning in the teaching profession. Providing structured mentorship programmes, where experienced educators guide newer teachers, can help balance traditional and modern instructional approaches in schools (Stolk et al., 2021).

Hypothesis Four:

Educator experience positively influences learner academic performance, but the effect diminishes at higher experience levels.

The findings support Hypothesis Four, demonstrating that educator experience positively influences learner academic performance but with diminishing returns at higher levels of experience; this shows the importance of balancing experience accumulation with continuous learning and instructional adaptation.

The results from OLS and RLS in Table 3 confirm that educator experience significantly positively affects learner academic performance, supporting the first part of Hypothesis Four. The estimated coefficient for experience is positive and statistically significant, indicating that increased teaching experience enhances student achievement. However, the coefficient for experience squared is negative and significant, confirming that the effect of experience on academic performance follows a diminishing returns pattern; this suggests that while initial years of experience contribute positively to student learning performance, the marginal benefits decline as experience accumulates (Mincer, 1974). The diminishing effect of experience on learner performance aligns with economic and educational theories suggesting that teaching effectiveness follows an inverted-U shape. Early in their careers, educators gain skills, refine pedagogical strategies, and develop classroom management techniques, leading to improved student performance. However, additional years of experience may contribute less significantly to learning performance beyond a certain threshold due to factors such as routine complacency, resistance to new teaching methods, or reduced professional development engagement (Bai & Zang, 2025). This finding is consistent with previous studies that stressed experience as a key determinant of student success, particularly in educators’ early- and mid-career phases (Stolk et al., 2021). However, long-tenured educators may require continuous professional development to maintain effectiveness and adapt to evolving curricula, technological advancements, and changing student needs.

The results indicate the need for ongoing training programmes, particularly for mid- and late-career educators, to ensure they continue enhancing their teaching methodologies rather than relying solely on accumulated experience. Policymakers should implement workshops and skill-refreshing courses to sustain high teaching effectiveness across all career stages. Also, the positive impact of experience, particularly in the early- and mid-career stages, suggests that mentorship programmes could be an effective strategy. Pairing experienced educators with newer teachers can help transfer best practices while introducing experienced teachers to newer, innovative teaching strategies (Stolk et al., 2021). Since diminishing returns indicate that very experienced educators may become less effective over time, education policies should encourage innovation among senior teachers. Providing incentives for engagement in research-based teaching methods, technology integration, and peer learning programmes can help counteract stagnation and sustain high educational performance (Bai & Zang, 2025).

Hypothesis Five:

Educator qualifications influence learner academic performance more than demographic characteristics (age, gender, and experience).

The findings reject Hypothesis Five, indicating that educator qualifications do not influence learner academic performance more than demographic characteristics. Instead, experience and gender play a more significant role, emphasising the importance of practical teaching skills, adaptability, and long-term professional development over formal academic credentials alone.

The empirical results in Table 3 provide mixed evidence regarding hypothesis five. The findings indicate that educator qualifications are not statistically significant in explaining learner academic performance, while demographic characteristics, particularly experience and gender, show significant effects; this suggests that contrary to the hypothesis, demographic factors exert a stronger influence on student performance than formal qualifications (Stevanović et al., 2021; Even & BenDavid-Hadar, 2025). Specifically, the coefficient for experience is positive and significant, confirming that teaching experience enhances academic performance. However, the negative and significant coefficient for experience squared indicates diminishing returns at higher experience levels. Additionally, educator gender is significant, implying that differences in teaching effectiveness or interaction styles between male and female educators may play a role in influencing student performance. However, age is not a significant predictor, suggesting that experience, rather than chronological age, drives effectiveness in the classroom.

The insignificance of educator qualifications challenges traditional assumptions that higher formal education directly translates into improved teaching effectiveness. While qualifications provide foundational pedagogical knowledge, they may not fully capture practical teaching skills, adaptability, and student engagement techniques, all of which are essential for academic success; this aligns with studies suggesting that on-the-job experience, professional development, and classroom management skills often have a greater impact on student outcomes than formal credentials alone (Tatto, 2021; Sims & Fletcher-Wood, 2021). Since experience and gender (possibly linked to teaching styles) significantly affect student performance, policymakers should balance credential requirements with practical teaching assessments when hiring educators; this may involve teaching demonstrations, probationary evaluations, and peer-reviewed classroom observations. The results indicate a need for a stronger integration of classroom-based training, mentorship, and continuous professional development into qualification programmes. Simply obtaining a degree may not be sufficient; institutions should emphasise interactive, hands-on pedagogical strategies. Likewise, since teaching experience significantly impacts learner performance, investment in early-career mentorship programmes, peer collaboration, and ongoing pedagogical training is crucial. Schools should implement structured professional development pathways to maximise the effectiveness of both novice and experienced educators.

5. Conclusions and Policy Recommendations

The influence of educators’ demographic characteristics on learner academic performance has garnered significant attention, particularly in developing countries. However, gaps remain in understanding how these factors, alongside professional qualifications, influence student performance, especially in under-resourced regions. This study examines the impact of educator demographics and qualifications on learner performance in five secondary schools in Amathole East, Eastern Cape, South Africa. The study clarifies the relative importance of demographic attributes versus professional training, informs teacher recruitment and development policies, and enriches the empirical literature aligned with global educational goals. The following objectives are investigated in the study: analyse the relationship between educator qualifications and learner academic performance; evaluate the effect of educator demographic factors (gender, age, and experience) on learner academic performance; compare the relative influence of demographic characteristics and professional qualifications on learner academic performance; and assess the validity of Mincer’s Earnings Function in capturing diminishing returns to experience. The study employs Ordinary Least Squares regression to achieve these objectives, with Robust Least Squares used for validation. The study uses data from 70 educators across five secondary schools from the SA-SAMS database.

The findings reveal that teaching experience and gender significantly influence learner performance, whereas age and qualifications exhibit no statistical effect. The results further show diminishing returns to experience. The findings suggest that while teaching experience enhances learner performance, its impact diminishes over time, indicating that additional years of experience beyond a certain threshold do not yield proportional gains in student performance. The significant effect of male and female educators on learner performance shows that differences in teaching styles, communication approaches, or subject specialisations between male and female teachers may influence academic achievement. The importance of understanding gender effects in the classroom and ensuring that teacher recruitment and professional development programmes are designed to influence the strengths of both male and female educators is stressed by this study. Conversely, the lack of statistical effect for age and formal qualifications challenges conventional assumptions that older or more credentialed educators necessarily contribute to better student performance; this outcome emphasises the need for teacher development programmes that provide continuous pedagogical improvement rather than solely relying on tenure or academic credentials.

Based on the findings, the need for education policies that prioritise teaching effectiveness over traditional indicators such as age or formal qualifications is recommended. Since experience significantly influences learner performance but exhibits diminishing returns over time, policymakers should consider implementing continuous professional development programmes. These programmes should focus on innovative teaching methodologies, student engagement strategies, and adaptive learning approaches to ensure that experienced educators continue to improve their skills and remain effective throughout their careers. Instead of relying solely on years of service, performance-based assessments and training incentives should be integrated into teacher development frameworks. The significant influence of gender on learner performance indicates the importance of considering gender impact in teacher recruitment and professional development policies. Differences in teaching styles, communication methods, and subject specialisations between male and female educators may affect student achievement. Therefore, education policies should promote gender diversity in teacher placement across subjects and grade levels to optimise learning experiences. Additionally, training programmes should be designed to strengthen male and female teachers and ensure that pedagogical strategies are inclusive and have diverse learning needs.

Furthermore, the absence of a statistical effect for age and formal qualifications suggests that policies should shift away from emphasising tenure or advanced degrees as primary indicators of teaching quality. Instead, education systems should prioritise practical teaching competencies and ongoing skill development. Teacher recruitment and promotion criteria should incorporate evidence-based assessments of classroom performance, student engagement effectiveness, and adaptability to modern educational demands; this shift would ensure that hiring and career progression decisions are based on merit and demonstrated teaching effectiveness rather than traditional credentials alone.

This study makes contributions by providing empirical evidence on the impact of educator demographics and qualifications on learner performance in under-resourced schools. However, its scope is limited by the small sample size and geographic focus, which may affect the macro applicability of the findings. Additionally, the reliance on secondary data excludes qualitative aspects of teaching effectiveness, such as instructional methods and classroom engagement. While the study pointed out the significant role of gender and experience in influencing academic performance, it does not examine the underlying mechanisms driving these effects. Moreover, external factors like school infrastructure and socioeconomic conditions were not accounted for, which could influence learner performance. Future research should build on this study by expanding the sample size, incorporating mixed-method approaches, examining gender-based teaching effects, and integrating contextual factors to provide a more comprehensive understanding of the determinants of student success. Likewise, future studies should investigate how differences in teaching styles, communication methods, and subject specialisations between male and female educators affect academic performance.