Enhancing Community Waste Recycling in Taiwan: Key Drivers Affecting Consumers in Waste Recycling

Abstract

The municipal solid waste generation is projected to spike from 2.1 billion tonnes in 2023 to 3.8 billion tonnes by 2050. In Taiwan, the upsurge of waste volume, in addition to periodic maintenance of incinerators, which may persist up to four months, has resulted in limited incineration capacity. The optimum approach to address the challenge is to reduce the amount of waste sent for incineration by effective segregation of combustible and non-combustible waste, as well as improving the public recycling rate. Local authorities play a significant role in encouraging public recycling and restricting non-burnable waste from being delivered to incinerators within a short period of time. This can greatly reduce the amount of waste and incinerator maintenance costs. This study aimed to explore the key driving factors for public participation in waste recycling and translate the determinants into policy in order to increase the waste recycling rate. The study employed literature analysis to select factors repeatedly mentioned as indicators and conducted online surveys to collect data on factors influencing consumer engagement in waste recycling in Taiwan. This study also adopted the Analytic Hierarchy Process and established a hierarchical framework with four dimensions (Psychological, Knowledge, Policy, and Infrastructure) and thirteen indicators. The findings have demonstrated that infrastructure (0.275) is the most influential aspect in affecting consumers’ recycling actions, followed by psychological (0.256) and policy aspects (0.251), and knowledge aspect (0.218) as the least influential factor. Positive rewards (0.120), recycling knowledge (0.118), and well-built infrastructure (0.113) were specifically identified as key drivers in encouraging recycling. The findings informed the public’s priorities in recycling involvement, and strategic initiatives targeted at these preferences can effectively assist local authorities in promoting citizen engagement in recycling. Policies that meet public demands, such as positive rewards for recycling, dissemination of recycling knowledge, and provision and improvement of more recycling infrastructure, can ensure the success of the policy implementation and serve as a reference for other Asian countries in reducing waste and improving the recycling rate.

1. Introduction

The fast-paced development of science and technology innovation, coupled with economic growth and expanding consumption capacity, has significantly accelerated waste generation. According to Kaza and her colleagues, approximately one-third of the world’s annual average of 2.01 billion tons of municipal solid waste (MSW) is not managed in an environmentally appropriate manner [1]. A report published in 2024 by the International Solid Waste Association (ISWA) and the UNEP (UN Environment Programme) also anticipated that the amount of MSW generated will increase to 3.8 billion tons by 2050 [2]. The global outbreak of the COVID-19 crisis in 2020 has brought about major shifts in people’s lifestyle, particularly in dietary habits and shopping patterns, leading to a surge of MSW from food and goods packaging. In Taiwan, for instance, waste from disposable cutlery and food packaging rose sharply from 90,000 tons in 2019 to 170,000 tons in 2020 [3]. Meanwhile, individuals’ purchasing patterns have shifted from physical stores to online shopping, increasing the use of packaging cartons from 2.36 million to 2.64 million metric tonnes. These lifestyle changes have caused an immense amount of waste to be generated, implying that individual behavioural change is extrinsically linked to growing waste pressure.

Despite the fact that there has been a persistent recycling culture since the mandatory recycling legislation in Taiwan, the Environmental Protection Administration (EPA)’s study in 2018 pointed out that garbage accumulation has increased by 261% over five years due to an upsurge in solid waste volume. The number of cleanfill sites has been rising annually to accommodate the growing volume of MSW, largely due to the limited capacity and prolonged maintenance period of incinerators, as well as the dearth of waste treatment facilities. The data from the EPA in 2023 also showed that the total volume of general waste in Taiwan hit a new high record of 9.87 million tons in 2020, with over half of the MSW generated being recyclable [3]. The EPA also highlighted that recycling plastic containers not only conserves energy but also yields significant economic benefits. For instance, recycled plastic containers can save approximately 760 million TWD annually in waste removal and disposal costs, while producing around 169,000 metric tons of recycled materials. This, in turn, generates an output value of about 4.83 billion TWD [4]. Therefore, boosting public involvement in waste segregation and recycling is imperative to both reducing the amount of waste delivered to incinerators or landfills and maximizing the value of recyclable resources.

MSW can be categorized into four main sources: household waste, agricultural waste, commercial waste (schools, shops, offices, etc.), and industrial waste. Of these, household solid waste constitutes the greatest proportion of MSW [5]. Although Taiwan has long encouraged recycling and garbage sorting, the growing volume of MSW necessitated intensified recycling efforts. Strengthening these efforts is essential to maximizing the societal benefits, conserving natural resources, reducing environmental pollution and greenhouse gas emissions, and minimizing the amount of waste requiring incineration.

This work proposes to first comprehend the key dimensions and indicators that affect public participation in recycling through a literature review, then examine the critical factors that influence Taiwan’s consumers in recycling engagement through online surveys and the analytical hierarchy process (AHP) for analysis.

The following sub-section reviews relevant publications on MSW management and the factors that influence people’s engagement in recycling, as well as outlines the research objectives. Section 2 describes the design of an online survey for data collection and the adoption of AHP in data analysis. Section 3 highlights the findings of this study, while Section 4 discusses the key factors affecting recycling behaviours across different demographic groups and offers policy-oriented recommendations. Finally, Section 5 concludes the study, acknowledges its limitations, and proposes directions for future research.

1.1. Literature Review

1.1.1. Municipal Solid Waste (MSW) Management

Waste has been identified as a major contributor to global climate change due to its substantial greenhouse gas emissions [6]. Ramachandra and his colleagues believed that the sharp rise in waste volume is primarily attributable to improper management of solid waste, inadequate implementation of 3Rs (Recycle, Reuse, Reduce), and inefficiencies in recycling and waste transportation [7]. While the issue of MSW management is daunting, a previous study pointed out that only 0.69% of publications revolved around this topic [8]. Knickmeyer believed that the research on waste management concerns on the social level is conspicuously lacking, with most studies still being geographically limited despite growing scholarly interest [9].

Today, current research on MSW management focuses mainly on the application of the waste-to-energy process and its impacts on the environment, society, and national economy. According to Ma and Hipel, the predominant variables influencing urban solid waste management systems at the societal level include: vulnerability, public participation, public attitudes and behaviours, and policies [8]. Thomas’s research further supported this statement by highlighting that public involvement can increase the effectiveness of the government’s recycling initiatives [10].

Among the publications that discussed public participation in MSW management, Varotto and Spagnolli concentrated on examining psychological aspects, while Adu-Gyamfi et al. identified convenience, environmental consciousness, hedonic motivation, and benefit perception as key determinants of individuals’ participation in garbage sorting [11,12]. In order to examine the factors influencing public engagement in waste classification and recycling in Taiwan, it is essential to investigate the behavioral priorities and motivations of local communities in adopting sustainable recycling practices.

1.1.2. Role of Consumer in Recycling

Consumers play an important role in implementing waste classification as they generate waste through their purchasing decisions, consumption habits, and disposal practices. Despite this, there remains a lack of a systematic and multidimensional approach in investigating the factors that affect public involvement in waste classification. Moreover, research from the standpoint of consumers has not been as well discussed. Since recycling behaviour and barriers to participation can vary depending on target groups and regions, sociodemographic traits are also frequently employed as analytical variables to investigate people’s involvement in recycling waste.

Rousta et al. have discovered that gender, age, income, and education level are the demographic parameters that are most frequently studied in relation to household waste sorting behaviours, based on a literature review covering 39 publications [13]. Beyond demographic distinctions, key influencing factors are typically categorized into psychological variables, economic factors, and regulatory policies [14]. To ensure transparency and focus while allowing flexibility in synthesizing diverse themes across factors that may influence recycling behaviour, a systematic-inspired narrative review was conducted using keywords like “recycling behaviour”, “recycling participation and factors”, “factors affecting recycling behaviour”, “social factors” and “factors affecting participation in recycling”. The search was carried out using online academic databases such as ScienceDirect and Scopus, focusing on studies published between 2012 and 2022 to include both recent advancements and ongoing trends in the field. The screening process applied several inclusion criteria: (a) studies addressing primary influencing factors such as psychological, economic, policies, facility level, and knowledge; (b) open access availability; and (c) peer-reviewed journals. Based on these criteria, a total of 64 studies were selected for the final review.

This study examined the factors that influence people’s engagement in recycling through 64 relevant literatures on waste source classification and management, as demonstrated in

Table 1. Performance indicators of the four key dimensions derived from the literature review.

Dimensions	Indicators	Definition	Reference Number
Psychological	(1) Time consumed	Effort and time required to clean and sort recyclable items	[9,12,13,20,22,23,24,25,26,27,28,29,30,31,32,33,34,35]
	(2) Recycling habit	A mode of automatically recycling and sorting waste	[9,12,13,22,23,28,29,30,32,33,34,36,37,38,39]
	(3) Personal attitude	Have positive attitude and responsibility for waste classification	[9,13,26,27,29,30,32,33,34,36,40,41,42,43,44,45,46]
	(4) Other influences	Change own behaviour patterns due to other people influences	[9,13,22,27,30,32,33,34,35,37,43,44,47,48,49]
Knowledge and education	(5) Recycling knowledge	Ability to identify and classify recyclable items	[8,11,12,13,15,17,19,21,25,26,29,33,39,41,42,43,44,45]
	(6) Packaging labels	Complex packaging with single label affects recycling	[9,12,13,23,27,28,29,33,36,39,47,50,51,52,53,54,55]
	(7) Multimedia communication	Raise awareness on recycling through multimedia	[9,13,14,24,25,33,43,44,47,48,49,55,56,57,58]
Economy and policy	(8) Positive rewards	Rewards like money, prizes, vouchers, discount coupons, etc.	[8,10,11,12,16,21,26,28,33,35,38,39,44,50,51,52,53,54,55,57,59]
	(9) Monetary penalties	Penalties like taxes, fines, etc.	[9,32,33,36,43,56,60,61,62,63]
	(10) System recognition	Recognition of local authority recycling system	[9,14,24,27,33,35,36,40,43,44,46,51,61,64]
Infrastructure	(11) Basic infrastructure	Adequacy of recycling bins and recycling stations	[8,11,15,16,17,19,20,21,23,25,26,28,33,36,38,39,41,47,48,50,52,53,55,59,61]
	(12) Recycling distance	Distance to the nearest recycling facility	[9,12,21,23,27,28,29,33,57,61,65,66,67,68,69]
	(13) Storage space	Extra spaces for temporary storage of recycling items	[9,12,23,26,27,28,29,31,33,35,47,61,66]

. The review found that there are four primary dimensions that are frequently cited, covering psychology, recycling knowledge, infrastructure, economy, and policy. The Theory of Planned Behaviour (TPB) serves as a widely accepted framework for understanding and predicting human behaviour by emphasizing the roles of intention, attitude, subjective norms, and perceived behavioural control (PBC) [15]. Chen and Tung have discovered a strong positive correlation between people’s engagement in waste sorting and perceived behavioural control (PBC), which Bandura defined as a person’s perceived ability to perform a specific action—in this case, recycle [16,17]. Situational factors also shape psychological variables by either facilitating or hindering recycling behavior. These include the time required, recycling habits, attitudes toward recycling, societal influences, and the observed behaviors of others. These elements can affect people’s willingness and ability to participate in recycling. Research indicated that the type of individual behaviour determines the degree of intention-behaviour stability; hence, translating intention into action requires a certain level of behavioral control, supported by adequate knowledge, awareness, and access to resources [15,18,19]. In addition, previous studies highlighted that the key to improving the efficiency of waste recycling depends on sufficient civic awareness. Therefore, educational initiatives—such as incorporating recycling information on packaging labels and through multimedia campaigns—are essential to increasing public knowledge and participation [20,21]. In addition, policies that can foster a conducive environment for recycling, along with accessible and efficient infrastructure, play a crucial role in shaping public intention and attitude towards recycling behaviour.

To ensure that the indicators used in the questionnaire are influential and relevant, only those referenced in at least ten of the reviewed publications were selected. As shown in Table 1, the final set included 13 key indicators distributed across the four identified dimensions, all of which are frequently discussed in the literature.

1.1.3. AHP in Waste Management

The Analytic Hierarchy Process (AHP) has been widely used in waste management studies. Previous research has employed this method to improve solid waste disposal strategies and identify the best locations for recycling facilities [70,71]. Moreover, Shahrasbi et al. have adopted a combination of quantitative and qualitative methods, based on social media, literature review, and AHP, to analyse consumer participation in e-waste recycling programs from a sustainable development perspective [72]. Since the factors that influence public engagement in recycling include multiple aspects, AHP is particularly well-suited for this study, as it enables the simultaneous evaluation of multiple criteria and alternatives. Drawing upon methodologies used in related research on solid waste recycling and management, this study adopted the AHP technique to analyze the consumers’ prioritization of factors influencing their participation in recycling. A detailed explanation of the AHP methodology applied in this research is provided in Section 2.1.

1.2. Research Objectives

The literature review revealed several research gaps: (a) limited research on municipal solid waste management; (b) little focus on social level of solid waste management system, specifically public participation; (c) lack of a systematic and multidimentional approach in investigating the factors that affect public involvement in waste classification; and (d) an underrepresentation of consumer’s role in waste classification practices. To fill in the research gaps and improve the waste recycling rate in Taiwan, this study seeks to investigate factors that influence consumers’ involvement in waste recycling. The objectives of the study include:

• To identify and evaluate the influential factors and indicators that encourage Taiwan consumers’ involvement in recycling using AHP
• To produce policy insights based on Taiwan consumers’ priorities in enhancing the effectiveness of the waste recycling system in Taiwan

This paper aimed to contribute to the improvement of the recycling system and increase the recycling rate in Taiwan by promoting public participation and informing the design of recycling policies that are better aligned with consumer needs and behaviours. The conceptual framework of this study is as displayed in Figure 1. This study incorporates 13 indicators across four primary dimensions to identify the key drivers of public recycling behavior, which can be translated into actionable policy recommendations.

2. Materials and Methods

This study employed a quantitative research method comprising an online survey for data collection and AHP for data analysis. The survey, designed using Google Forms, was disseminated across Taiwan via social media platforms. It targeted consumers aged 21 and above, encompassing a broad demographic with purchasing capability. Unlike many AHP-based studies that rely on expert panels, this research engaged the general public as respondents to better capture the actual preferences and priorities of consumers. This approach reduces the risk of expert-driven bias and avoids a top-down perspective that may not accurately reflect the true needs of the public. The online survey was conducted from October 2023 to November 2023 and collected 140 responses in total. After filtering and eliminating incomplete survey forms, a total of 110 valid questionnaires were used as the empirical data in this study. The respondents were informed that their participation was entirely voluntary, and consent was obtained from all subjects involved in the study.

Apart from evaluating the importance of the 13 performance indicators related to recycling participation, the questionnaire also included seven demographic variables: gender, age, education level, occupation, income range, family participation in recycling, and individual participation in recycling. These variables enabled an examination of how demographic characteristics influence public participation in recycling practices.

To assess the representativeness of the sample, demographic data from the respondents were compared with national population statistics obtained from Taiwan’s Ministry of the Interior. The gender distribution of the sample—49.1% male and 50.9% female—closely mirrored the national figures of 49.4% male and 50.6% female. The age distribution of respondents is as demonstrated in Figure 2. The respondents from 21–30 years old made up the majority (30.9%), followed by age group 31–40 years old (24.5%), 51–60 years old (20.0%), 41–50 years old (14.5%), with age group above sixty years old (10%) as the least. Compared to national demographics, the sample skewed toward younger and middle-aged individuals, with an underrepresentation of respondents aged over 60.

Geographically, the respondents of this study are found to be primarily concentrated in the central region of Taiwan (47.3%), followed by the northern (31.8%), southern (19.1%), and eastern regions (1.8%) as displayed in Figure 3. As the survey was conducted online, disparities in digital literacy and access to information and communications technology (ICT) may have affected participation—particularly among older adults who may be less familiar with digital devices.

2.1. Analytical Hierarchy Process (AHP)

Among the literature discussing solid waste management, Multiple Criteria Decision Making (MCDM) is frequently employed, with the AHP being the most commonly used technique. According to Soltani et al., approximately 65% of the literature that addressed solid waste management using MCDM has adopted AHP as the primary analytical tool [73]. It is frequently employed to explore management strategies and influencing factors in order to strengthen and improve solid waste recycling. One of its key advantages lies in its ability to deconstruct complex decision problems into a structured hierarchical model, thereby clarifying the relationships among evaluation criteria.

By calculating the relative weights of indicators through pairwise comparisons, AHP provides a sound basis for evaluation and decision-making. The use of eigenvalues from the pairwise comparison matrix enables consistency checks, which enhance the reliability and validity of the results. Since improving recycling rates typically requires the consideration of multiple interrelated criteria, four of which were identified in this study, AHP is well-suited for this type of analysis.

Additionally, AHP was selected in this study because it allows for the identification and prioritisation of the most influential factors by comparing criteria against one another, ensuring that all criteria are properly considered and weighed. While there are 13 alternative indicators used in this study, these are grouped and compared against each other within each dimension to simplify the pairwise comparison process. As such, AHP is adopted in this study to investigate the most influential dimension and preferred alternatives in encouraging public participation to improve the waste recycling rate in Taiwan.

The research process has been illustrated in Figure 4 and can be divided into the following five steps.

1.

Problem definition

The first step involves identifying the key factors influencing the problem through a literature review. This study focused on determining the variables affecting public participation in recycling and formulating potential strategies for improvement.

2.

Establishing a hierarchical framework

At this stage, a hierarchical framework is constructed by breaking down the problem and influential factors into a series of levels. The top level represented the overall objective—improving Taiwan’s waste recycling rate. This is followed by four main criteria at the second level, derived from a review of 64 relevant studies: psychological factors, knowledge and education, economic and policy influences, and infrastructure. At the third level, 13 alternative indicators identified from the literature were included as potential factors promoting public participation in recycling (see Figure 5).

3.

Questionnaire design and survey

The questionnaire targeted the general public, aiming to capture a broad range of consumer perspectives. To ensure consensus among participants in AHP using an online survey, the questionnaire is carefully structured with clear definitions of each criterion, comprehensive instructions supported by the AHP hierarchical framework, as well as a consistent scale for pairwise comparisons. A pilot test was conducted with a small group of acquaintances (family and friends) to refine the clarity and usability of the questionnaire before its broader distribution. The final version is administered online via Google Forms.

The questionnaire is designed to first compare the 4 criteria against each other using a nominal scale. The respondents are then required to compare the alternatives against each other within each criterion group, as shown in Figure 6. The comparisons were based on a standardized nine-point scale ranging from “equally important” (value = 1) to “absolutely more important” (value = 9), as portrayed in

Table 2. The evaluation scale and explanation of the analytic hierarchy process (AHP).

Evaluation Scale	Definition	Explanation
1	Equal Importance	The importance of two factors and degree of influence is equally strong
3	Moderate Importance	The importance slightly inclined to a certain factor and the degree of influence is slightly stronger
5	Essential Importance	The importance strongly inclined to a certain factor and the degree of influence is strong
7	Very Strong Importance	The importance very strongly inclined to a certain factor and the degree of influence is very strong
9	Absolute Importance	The importance must be absolute to a certain factor and the degree of influence is extremely strong
2, 4, 6, 8	Intermediate values between two adjacent judgements	Use when compromise is needed

Source from: [74].

. The weightage allocated to each criterion and alternative depends on the respondent’s subjective judgement regarding the relative importance of each factor influencing recycling participation. This structure enabled the aggregation of individual preferences and the derivation of priority weights for each factor, consistent with the principles of AHP. The questionnaire was designed in Chinese but has been translated into English and is presented in Appendix A.4.

4.

Weightage calculation

The calculation part can be divided into two steps: (1) Creation of a pairwise comparison matrix, (2) Calculation of eigenvalues and eigenvectors. By comparing each factor at the same level in pairs, the pairwise comparison matrix’s value is derived from the previous stage using a questionnaire survey. After the pairwise comparison matrix is established, the eigenvalues and eigenvectors are calculated using Microsoft Excel.

• Establishment of pairwise comparison matrix: The values used for pairwise comparison are obtained from the questionnaire survey Each respondent compares factors at the same hierarchical level, and their judgments are entered into the upper triangular part of the comparison matrix. By definition, the diagonal elements of the matrix are set to 1, as they represent the comparison of a factor with itself. The lower triangular elements are the reciprocals of the corresponding upper triangle values, such that $aji=1/aij$. The general form of the pairwise comparison matrix A is:

  $$\mathrm{A}=\left[{\mathrm{a}}_{\mathrm{ij}}\right]=\left[\begin{array}{cccc}1& {\mathrm{a}}_{12}& \dots & {\mathrm{a}}_{1\mathrm{n}}\\ {\mathrm{a}}_{21}& 1& \dots & {\mathrm{a}}_{2\mathrm{n}}\\ ⋮& ⋮& \ddots & ⋮\\ {\mathrm{a}}_{\mathrm{n}1}& {\mathrm{a}}_{\mathrm{n}2}& \dots & 1\end{array}\right]=\left[\begin{array}{cccc}1& {\mathrm{a}}_{12}& \dots & {\mathrm{a}}_{1\mathrm{n}}\\ {\displaystyle \frac{1}{{\mathrm{a}}_{12}}}& 1& \dots & {\mathrm{a}}_{2\mathrm{n}}\\ ⋮& ⋮& \ddots & ⋮\\ {\displaystyle \frac{1}{{\mathrm{a}}_{1\mathrm{n}}}}& {\displaystyle \frac{1}{{\mathrm{a}}_{2\mathrm{n}}}}& \dots & 1\end{array}\right]$$

  (1)
• Eigenvalue and eigenvector calculation: Once the matrix is constructed, the eigenvector representing the relative weights of the criteria or alternatives is calculated. In this study, Microsoft Excel was used for computation. The row vector average standardisation method is typically employed when calculating vector values in the AHP, and the calculation accuracy is higher [74]. Method of standardising row vector averages:

  $${\mathrm{W}}_{\mathrm{i}}^{\prime }={\displaystyle \frac{1}{\mathrm{n}}}{\sum }_{\mathrm{j}=1}^{\mathrm{n}}{\displaystyle \frac{{\mathrm{a}}_{\mathrm{ij}}}{{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{a}}_{\mathrm{ij}}}} \mathrm{i}, \mathrm{j}=1, 2, \dots \dots ,\mathrm{n}$$

  (2)

5.

Consistency test

To ensure that the respondents’ judgments are logically consistent, AHP includes a consistency check. This involves the calculation of the Consistency Index (C.I.) and the Consistency Ratio (C.R.), which assess the reliability of the pairwise comparisons.

• Consistency Index, C.I.

To test the consistency between before and after values, the calculation of the consistency index was performed. The formula to calculate Consistency Index (C.I.) is as follows:

$$\mathrm{C}.\mathrm{I}.={\displaystyle \frac{\mathsf{\lambda }-\mathrm{n}}{\mathrm{n}-1}}$$

(3)

In order to determine the λ value, which is a prerequisite for determining the C.I., the weight w that was previously determined to discover the consistency vector (represented by v) was used. The equation is:

$${\mathsf{\nu }}_{\mathrm{i}}=({\sum }_{\mathrm{j}=1}^{\mathrm{n}}{\mathrm{w}}_{\mathrm{j}}{\mathrm{a}}_{\mathrm{ij}})/{\mathrm{w}}_{\mathrm{i}} \mathrm{i}, \mathrm{j}=1, 2, \dots \dots , \mathrm{n}$$

(4)

After obtaining the consistency vector, λ value is obtained by finding the arithmetic mean of its v values through the formula:

$$\mathsf{\lambda }={\displaystyle \frac{{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\mathsf{\nu }}_{\mathrm{i}}}{\mathrm{n}}} \mathrm{i}=1, 2, \dots \dots , \mathrm{n}$$

(5)

Evaluation criteria:

When the C.I. value < 0.1, it means the consistency is within the acceptable range.

When the C.I. value = 0.1, it means the judgments before and after are completely consistent.

When the C.I. value > 0.1, it means the consistency of the judgments before and after is low. Inconsistencies should be found, adjusted, and corrected until the consistency passes the test.

• Consistency Ratio, C.R.

The Consistency Ratio (C.R.) indicates how w,a decision-maker’s judgments, align with the transitivity principle. The R.I refers to the random index (R.I.) of the matrix, and is as shown in

Table 3. Random Index (R.I.) value.

n	1	2	3	4	5	6	7	8	9
R.I.	0	0	0.58	0.9	1.12	1.24	1.32	1.41	1.45

Source from: [74].

. The equation to calculate C.R. is as shown below:

$$\mathrm{C}.\mathrm{R}.={\displaystyle \frac{\mathrm{C}.\mathrm{I}.}{\mathrm{R}.\mathrm{I}.}}$$

(6)

Evaluation criteria:

When the C.R. value is ≤0.1, it means that the consistency has reached an acceptable level.

When the C.R. value is >0.1, it means that the consistency is not met and adjustments are required.

2.2. Statistical Test

This study employed statistical software IBM SPSS version 29.0.2.0 to explore whether demographic characteristics influence the prioritization of dimensions and indicators derived from the AHP analysis. A Kolmogorov-Smirnov (K-S) test is used in this study to assess whether the data followed a standard normal distribution. Given the non-parametric nature of the data, the Mann–Whitney U test was applied to determine whether significant differences in indicator weightings exist across gender, family influence on recycling behavior, and individual recycling habits. Additionally, the Kruskal–Wallis test was employed to evaluate differences in weightings based on respondents’ age, education level, and income.

Mann-Whitney U test is a non-parametric statistical test that compares two independent samples and is an alternative to the Student’s t-test. It is useful when data are not ordinal, not normally distributed, or when the assumptions of parametric tests like the t-test are violated. This study employed the Mann-Whitney U test to determine whether significant differences exist in factor weightage based on (a) respondent’s gender, (b) whether the respondent personally adopts recycling habits, and (c) whether the respondent’s originating families influence their recycling habits.

Meanwhile, the Kruskal-Wallis test is a non-parametric statistical test used to determine whether there are statistically significant differences between three or more independent groups on a continuous or ordinal variable. It is the non-parametric alternative to one-way ANOVA and is particularly useful when the normality or homogeneity of variances is violated. This study applied Kruskal–Wallis test to evaluate differences in factor weights across consumer age, education level and income ranges.

3. Results

The present study aimed to investigate the key factors influencing the consumer’s participation in recycling in Taiwan. Through an online survey, three key drivers emerged as most influential: recycling knowledge, a positive reward system, and basic recycling infrastructure. These three factors were found to be the priorities of consumers that motivate them to engage in recycling behaviour.

Based on the empirical data collected through online questionnaires, the four key dimensions with thirteen performance indicators were evaluated for their importance in influencing consumers’ participation in recycling using a nominal scale from 1 to 9. The calculations of the weight of the key dimensions and indicators were conducted using Microsoft Excel and applied the AHP methodology. The dimension and indicator with a higher value indicate that the factor has a stronger influence on consumers’ participation in recycling behaviour. These findings informed the consumer priorities for policy development aimed at enhancing Taiwan’s recycling rate through increased public engagement.

3.1. Factors Influencing Consumer Recycling

The four major dimensions have been compared in pairs to explore the preference of respondents regarding factors influencing recycling behaviour. Among these, the Infrastructure dimension scored the highest weightage of 0.275, followed by Psychological (0.256), Economy and policy (0.251), and finally the Knowledge and education dimension (0.218). A consistency test has been performed to validate the reliability of the results, and the C.R. value was calculated using equation 6 as stated in Section 2, Materials and Methods. The C.R. value obtained was 0.0381 (≤0.1), indicating that the consistency has reached an acceptable level. The findings suggested that the availability and quality of recycling infrastructure are the most influential factors in motivating customers to recycle. Meanwhile, the close scores among the psychological, economic, and policy aspects imply that they also have an equivalent influence on consumer recycling behaviour.

At the indicator level, it is distinct that recycling habits (Psychological), recycling knowledge (Knowledge and education), positive reward (Economy and policy), and basic infrastructure (Infrastructure) received the highest weightage in their respective dimensions. Within the Psychological dimension, recycling habit had the highest score of 0.335, ahead of time consumed, personal attitude, and other influences. In the Knowledge and Education dimension, recycling knowledge emerged as the most important factor (weight = 0.472), nearly double the score of multimedia communication and packaging labels.

Similarly, within the Economy and Policy dimension, the indicator positive reward (weight = 0.48) ranked significantly higher than monetary penalty and system recognition, suggesting that positive stimulus, which can provide additional gain and foster a supportive environment, is more motivating in behavioural change. In the Infrastructure dimension, the availability of basic infrastructure such as recycling bins, recycling stations, and other recycling facilities received the highest weight (0.45), surpassing recycling distance and storage space as key considerations.

Overall weightage was then calculated to obtain the overall weightage of all indicators. When the indicators’ overall weightage is compared across 4 dimensions, the Economics and Policy dimension’s positive reward achieved the highest overall weighting of 0.120, followed closely by recycling knowledge (0.118) and basic infrastructure (0.113). On the other hand, others influence from the Psychological dimension received the lowest overall weight, suggesting that societal influence plays a relatively minor role in encouraging public participation to recycle. Despite that, the Knowledge and Education dimension received the lowest weightage among the four major dimensions, indicator recycling knowledge ranked second overall across all indicators, indicating that consumers also perceived its importance in fostering greater public participation.

The detailed weight analysis of the key factors influencing solid waste recycling is presented in

Table 4. Weightage evaluation analysis of key factors for solid waste recycling improvement.

Dimensions	Weightage (A)	Indicators	Weightage (B)	Overall Weightage (A × B)
Psychological	0.256	Time consumed	0.212	0.053
		Recycling habit	0.335	0.084
		Personal attitude	0.293	0.073
		Others influence	0.159	0.040
Knowledge and Education	0.218	Recycling knowledge	0.472	0.118
		Packaging label	0.282	0.071
		Multimedia communication	0.245	0.061
Economy and Policy	0.251	Positive reward	0.480	0.120
		Monetary penalty	0.280	0.070
		System recognition	0.240	0.060
Infrastructure	0.275	Basic infrastructure	0.450	0.113
		Recycling distance	0.320	0.080
		Storage space	0.230	0.057

Note: C.R. = 0.0381.

.

3.2. Demographic Characteristics That Influence Consumer Involvement in Waste Recycling

According to previous studies, demographic variables are found to predominantly influence the public’s engagement in garbage recycling [13]. Among these, age, income, and educational level were found to have a significant influence on recycling behaviours.

3.2.1. Gender

Among the 110 valid responses collected, 95 respondents reported active participation in recycling. Of these, approximately 49% were male and 51% were female, indicating an even gender distribution. The statistical analysis revealed that there is no significant difference in indicator weightage between male and female respondents, indicating that the perspective of both genders towards these factors is rather similar. According to

Table 5. Comparison of the weightage evaluation of indicators between genders.

Dimensions	Performance Indicators	Male	Female
Psychological	Time consumed	0.058	0.049
	Recycling habit	0.084	0.083
	Personal attitude	0.069	0.078
	Others influence	0.039	0.040
Knowledge and Education	Recycling knowledge	0.119	0.117
	Packaging label	0.071	0.071
	Multimedia communication	0.060	0.062
Economy and Policy	Positive reward	0.131	0.109
	Monetary penalty	0.064	0.076
	System recognition	0.055	0.065
Infrastructure	Basic infrastructure	0.126	0.100
	Recycling distance	0.072	0.088
	Storage space	0.052	0.062

, male respondents were more inclined to positive rewards and basic infrastructure, whereas female consumers placed greater emphasis on recycling knowledge. This implies that while overall gender influence may be balanced, specific motivational drivers vary slightly by gender.

3.2.2. Age

Figure 7 illustrates that respondents aged above 60 years old were the age group that participated the least in recycling (4.5%), while the younger generation aged from 21–30 years old made up the most (26.4%). These findings highlight generational differences in recycling engagement, possibly influenced by lifestyle habits, awareness, and accessibility to recycling facilities.

Table 6. Comparison of the weightage evaluation of indicators among different age groups.

Dimensions	Performance Indicators	Age Group (Years Old)
		21–30	31–40	41–50	51–60	Above 60
Psychological	Time consumed	0.043	0.048	0.071	0.042	0.092
	Recycling habit	0.084	0.085	0.086	0.082	0.081
	Personal attitude	0.077	0.075	0.065	0.085	0.046
	Others influence	0.046	0.042	0.028	0.041	0.031
Knowledge and Education	Recycling knowledge	0.115	0.130	0.134	0.107	0.097
	Packaging label	0.075	0.061	0.063	0.078	0.075
	Multimedia communication	0.060	0.059	0.052	0.065	0.078
Economy and Policy	Positive reward	0.117	0.108	0.143	0.117	0.129
	Monetary penalty	0.068	0.079	0.065	0.061	0.077
	System recognition	0.065	0.063	0.042	0.071	0.044
Infrastructure	Basic infrastructure	0.127	0.097	0.113	0.116	0.101
	Recycling distance	0.075	0.088	0.080	0.069	0.098
	Storage space	0.048	0.065	0.057	0.065	0.051

demonstrates the indicator weightings by age group. Respondents aged 21 to 30 placed the highest weightage (0.127) on basic infrastructure, suggesting that convenience and availability of comprehensive recycling facilities are primary motivators for younger consumers. Respondents aged 31 to 40 prioritize recycling knowledge (0.130), indicating that understanding proper recycling practices plays a pivotal role in motivating this demographic. In contrast, respondents aged 41 and above were all more likely to be motivated by positive rewards, implying a greater sensitivity to tangible benefits as age grows.

3.2.3. Income

Figure 8 has demonstrated that there were more respondents with an income level above TWD 20,000 engaged in recycling, suggesting a potential link between financial stability and recycling engagement. As detailed in

Table 7. Comparison of the weightage evaluation of indicators among different income levels.

Dimensions	Performance Indicators	Income Level (TWD)
		No Income	Below 20,000	20,000–39,999	40,000–59,999	Above 60,000
Psychological	Time consumed	0.067	0.067	0.041	0.060	0.054
	Recycling habit	0.081	0.081	0.085	0.076	0.091
	Personal attitude	0.054	0.055	0.081	0.074	0.075
	Others influence	0.048	0.047	0.043	0.039	0.031
Knowledge and Education	Recycling knowledge	0.076	0.110	0.117	0.122	0.133
	Packaging label	0.067	0.065	0.069	0.079	0.069
	Multimedia communication	0.107	0.075	0.064	0.049	0.048
Economy and Policy	Positive reward	0.098	0.132	0.124	0.123	0.115
	Monetary penalty	0.097	0.073	0.060	0.068	0.074
	System recognition	0.055	0.046	0.066	0.058	0.061
Infrastructure	Basic infrastructure	0.094	0.122	0.111	0.120	0.112
	Recycling distance	0.101	0.081	0.079	0.067	0.085
	Storage space	0.055	0.047	0.059	0.063	0.054

, consumer income emerged as an important variable influencing recycling motivations, as people from different income levels demonstrated varied considerations and preferences in indicators. Respondents without income placed the greatest importance on multimedia communication, emphasizing the role of media in disseminating recycling information and raising awareness. This finding suggests that for individuals not engaged in formal employment—such as students or retirees—information dissemination via television, online platforms, and social media is a key driver of engagement. On the other hand, respondents with a moderate income range (TWD 20,000–TWD 59,999) found positive rewards to be appealing and motivating. This indicates that this group responds better to financial or material incentives that align with their economic circumstances. Recycling knowledge seems to be significant to respondents with incomes over TWD 60,000, suggesting that they prioritised the practical engagement in recycling efforts with sufficient expertise on the topic.

These findings imply that effective policy should ensure equitable access to recycling information for individuals across all income levels. Tailored strategies such as targeted multimedia campaigns and educational initiatives should be designed to engage diverse income groups and promote recycling participation.

3.2.4. Education Level

Figure 9 revealed that the respondents who recycle from the three education categories were distributed fairly balanced. According to

Table 8. Comparison of the weightage evaluation of indicators among different education levels.

Dimensions	Performance Indicators	Education Level
		Secondary Education	Diploma/Bachelor Degree	Postgraduate Degree
Psychological	Time consumed	0.054	0.049	0.058
	Recycling habit	0.088	0.080	0.086
	Personal attitude	0.069	0.080	0.069
	Others influence	0.038	0.041	0.037
Knowledge and Education	Recycling knowledge	0.109	0.122	0.128
	Packaging label	0.075	0.069	0.063
	Multimedia communication	0.066	0.060	0.059
Economy and Policy	Positive reward	0.127	0.125	0.100
	Monetary penalty	0.061	0.070	0.086
	System recognition	0.061	0.056	0.065
Infrastructure	Basic infrastructure	0.109	0.112	0.121
	Recycling distance	0.082	0.076	0.081
	Storage space	0.059	0.062	0.048

, respondents with both a secondary education (0.127) and a diploma/bachelor’s degree (0.125) perceived the rewards and incentives system as the most motivating factor in recycling involvement. These findings suggest that tangible benefits, such as financial incentives, play a critical role in motivating individuals with low to moderate educational attainment to adopt recycling behaviours.

In contrast, postgraduate degree holders rated recycling knowledge as the most critical metric (0.128). This reflects that individuals with advanced education have a higher tendency to value knowledge-based motivations for pro-environmental behaviour. These results highlight the importance of designing differentiated strategies in promoting recycling. Therefore, policy frameworks should consider educational diversity to ensure that the recycling initiatives are effectively targeted and inclusive.

3.2.5. Family Recycling Habits Influence on Personal Behaviour

In referring to Figure 10, only 3.6% of respondents did not recycle, even when their families recycled. Conversely, approximately 52% of respondents who grew up in families where recycling was practiced reported to engage in recycling themselves. These results align with family systems theory, which posits that family members are interdependent, can influence one another directly or indirectly, and their interactions are transactional [75,76]. Accordingly, a person’s actions and perspective towards recycling might be greatly influenced by their family members, family’s education, living environment, and the habits they have cultivated since childhood. Apart from this, there are also about 33.6% of respondents who participate in waste recycling even without family influence, implying that personal attitude, individual environmental awareness, or external motivators such as education, media exposure, or policy incentives may also shape recycling behaviour independently of family background.

Based on

Table 9. Comparison of the weightage evaluation of indicators among individuals with or without family influence in recycling.

Dimensions	Performance Indicators	Family Influence		Without Family Influence
		Recycle	Do Not Recycle	Recycle	Do Not Recycle
Psychological	Time consumed	0.045	0.058	0.074	0.073
	Recycling habit	0.086	0.083	0.082	0.060
	Personal attitude	0.081	0.071	0.054	0.046
	Others influence	0.039	0.038	0.040	0.072
Knowledge and Education	Recycling knowledge	0.114	0.130	0.106	0.103
	Packaging label	0.074	0.070	0.053	0.070
	Multimedia communication	0.062	0.050	0.090	0.076
Economy and Policy	Positive reward	0.120	0.114	0.127	0.156
	Monetary penalty	0.067	0.071	0.082	0.064
	System recognition	0.063	0.065	0.042	0.030
Infrastructure	Basic infrastructure	0.114	0.104	0.118	0.158
	Recycling distance	0.079	0.085	0.081	0.043
	Storage space	0.057	0.060	0.051	0.049

, those who recycle, regardless of whether or not with family influence, perceived positive reward as the most motivating factor. On the other hand, basic infrastructure was viewed as the most influential indicator for people who neither recycle nor come from recycling households, revealing the importance of convenience in encouraging initial behavioural change. Among respondents who reported not recycling despite family influence, recycling knowledge was ranked as the most influential factor for potential behavioural change. This highlights a possible knowledge gap that may hinder action despite positive family norms, suggesting that targeted educational efforts could play a crucial role in translating awareness into action.

3.3. Analysis

Prior to conducting statistical analyses, IBM SPSS version 29.0.2.0 is employed to test for data normality. A K-S test has been performed, and it was discovered that the sample data did not follow a standard normal distribution. Given the violation of normality assumptions, non-parametric tests were deemed appropriate. To determine whether the weights derived from the analytic hierarchy process (AHP) differed significantly across various demographic variables, the Mann-Whitney U test and Kruskal-Wallis test were employed.

3.3.1. Data Normality Test

This study used the Kolmogorov-Smirnov (K-S) test to assess the sample data normality. The assumptions of the K-S test are as follows:

• Null hypotheses (H₀): The sample distribution is the same as the theoretical distribution (i.e., normal distribution).

Alternative hypotheses (H₁): The sample distribution is different from the theoretical distribution (i.e., normal distribution).

The p-values obtained from the normality tests are as shown in Appendix A.1. This study performed routine statistical tests on the samples, covering the 13 indicators. At a 95% confidence level, the result suggested that none of the examined variables followed a normal distribution. Therefore, this study adopted the Mann-Whitney U test and the Kruskal-Wallis test for subsequent analysis and validation.

3.3.2. Mann-Whitney U Test

As a non-parametric statistical test, the Mann-Whitney U test is employed in this study to examine whether gender, family recycling influence, and personal recycling habit significantly influence their evaluation across various indicators.

A.

Gender

Hypotheses proposed to examine whether gender significantly influences the perceived importance of specific indicators:

H₀: 

${\eta }_1={\eta }_2$, There is no significant difference between males and females in the perceived importance of indicators.

H₁: 

${\eta }_1\ne {\eta }_2$, There is a significant difference between males and females in the perceived importance of indicators.

The results of the analysis are as presented in Appendix A.2. Based on an evaluation of thirteen indicators across gender groups, statistically significant differences were identified in three indicators: positive reward (p = 0.024 < 0.05), basic infrastructure (p = 0.010 < 0.05), and recycling distance (p = 0.045 < 0.05). As such, the null hypotheses (H₀) are rejected, indicating that gender has a significant influence on their perceived importance.

B.

Originating family influences on recycling behaviour

Hypotheses proposed to examine whether the recycling practices of respondent’s originating family significantly affect the perceived importance of specific indicators:

H₀: 

${\eta }_1={\eta }_2$, There is no significant difference in the importance of indicators between respondent whose families engaged in recycling and those who did not.

H₁: 

${\eta }_1\ne {\eta }_2$, There is a significant difference in the importance of indicators between respondent whose families engaged in recycling and those who did not.

As shown in Appendix A.2, the analysis revealed that no statistically significant differences are observed between the two groups for any of the indicators. Accordingly, the null hypothesis is accepted, suggesting that the recycling behaviour of an individual family do not significantly affect the perceived importance of these indicators.

C.

Personal recycling habit

Hypotheses proposed to assess whether an individual’s engagement in recycling behaviour significantly influences the importance assigned to various recycling indicators.

H₀: 

${\eta }_1={\eta }_2$, There is no significant difference in the importance of indicators between respondents whose families engaged in recycling and those who did not.

H₁: 

${\eta }_1\ne {\eta }_2$, There is a significant difference in the importance of indicators between respondents whose families engaged in recycling and those who did not.

According to the result shown in Appendix A.2, three indicators: time consumed (p = 0.0364 < 0.05), personal attitude (p = 0.00903 < 0.05), and system recognition (p = 0.0186 < 0.05) were found to have statistically significant differences. Therefore, the null hypotheses are rejected for these indicators, implying that respondents who engage in recycling behaviour assign significantly higher importance levels to these three indicators compared to those who do not recycle.

3.3.3. Kruskal-Wallis Test

This study employed Kruskal-Wallis test to analyse whether respondent’s age, education levels, and income significantly influence their evaluation across various indicators. This method is used to compare whether there are statistically significant differences in the medians between three or more groups.

A.

Age

Hypotheses proposed to examine whether age significantly influences the perceived importance of indicators.

H₀: 

There are no significant differences in the median values of the indicators among the age groups 21–30, 31–40, 41–50, and above 60.

H₁: 

At least one age group differs significantly in the median values of the indicator compared to the others.

An analysis of thirteen indicators across different age groups, as shown in Appendix A.3, revealed statistical differences in time consumed and personal attitude. Therefore, the null hypotheses were rejected, indicating that the influence of these indicators varies significantly across respondents’ age groups.

As shown in

Table 10. p-value Matrix from Post Hoc Analysis (Dunn’s Test) on the Relationship Between Age and Time Spent and Attitude Norms.

	Time Consumed
Age	21–30	31–40	41–50	51–60	Above 60
21–30	1	1	1	1	0.022004
31–40	1	1	1	1	0.029117
41–50	1	1	1	1	1
51–60	1	1	1	1	0.02575
Above 60	0.022004	0.029117	1	0.02575	1
	Personal Attitude
Age	21–30	31–40	41–50	51–60	Above 60
21–30	1	1	1	1	0.045252
31–40	1	1	1	1	0.115199
41–50	1	1	1	1	0.709448
51–60	1	1	1	1	0.02608
Above 60	0.045252	0.115199	0.709448	0.02608	1

, post hoc analysis using Dunn’s test further identified specific group differences. For time consumed, respondents aged over 60 were significantly more affected compared to other age groups. Regarding personal attitude, the group aged 60 and above exhibited significantly higher sensitivity than all other age groups.

B.

Education level

Hypotheses were proposed to analyse whether education level significantly influences the weights assigned to recycling indicators.

H₀: 

There are no significant differences in the importance assigned to the indicators among respondents with an educational background of secondary school, diploma/bachelor’s degree, and postgraduate degree.

H₁: 

At least one educational group differs significantly in the importance assigned to the indicators.

According to Appendix A.3, the results indicated no statistically significant differences in the perceived importance of any indicator (p ≥ 0.05). This suggests that respondents, regardless of their educational background, hold similar views regarding the importance of the evaluated indicators. Therefore, the null hypothesis (H₀) is accepted.

C.

Income

Hypotheses were proposed to determine if income level significantly affects the perceived importance of indicators.

H₀: 

There are no significant differences in the perceived importance of the indicators among respondents with no income and those with income levels below TWD 20,000, TWD 20,000–39,999, TWD 40,000–59,999, and above TWD 60,000.

H₁: 

At least one income group differs significantly in the importance assigned to the indicators.

The analysis, as portrayed in Appendix A.3, revealed a statistically significant difference in the importance assigned to the indicator multimedia communication. Consequently, the null hypothesis is rejected for this indicator. This indicates that the perceived influence of multimedia varies significantly among respondents without income and with varied income ranges.

As shown in

Table 11. Dunn’s test p-value matrix for multimedia communication across income groups.

	Multimedia Communication
Income (TWD)	No Income	Below 20,000	20,000–39,999	40,000–59,999	Above 60,000
No income	1	1	0.215073	0.029925	0.001781
Below 20,000	1	1	1	1	0.797636
20,000–39,999	0.215073	1	1	1	0.221793
40,000–59,999	0.029925	1	1	1	1
Above 60,000	0.001781	0.797636	0.221793	1	1

, post hoc analysis using Dunn’s test suggested that individuals without income were more significantly influenced by multimedia communication compared to those with higher income (TWD 40,000–59,999 and above TWD 60,000).

4. Discussion

The result of this study demonstrated that the willingness of consumers in Taiwan to participate in recycling activities is positively correlated with the increased level of recycling knowledge, provision of positive rewards, and accessibility to comprehensive recycling infrastructure. These results underscore the multifaceted nature of recycling behaviour and align with broader waste management literature emphasizing the role of both policy and economic instruments in shaping consumer behaviour [45]. Consistent with previous research, this research supports the statement that monetary rewards such as gifts, prizes, vouchers, and coupons are more effective than punitive measures in motivating people to participate in waste classification and recycling [12,16,46]. This indicated that creating a supportive environment and rewarding system that recognises consumers’ effort in recycling is imperative to increase public participation in recycling.

The study further highlights the significance of recycling infrastructure as the most critical factor influencing consumer participation. The results of this study also corroborated earlier studies that emphasized the importance of highly accessible and well-equipped recycling infrastructure in boosting recycling rates because people are more inclined to recycle when their demands are satisfied [57,65,66]. In Taiwan, the current recycling framework comprises various channels, including private recycling stations that offer monetary returns for recyclable materials, and municipal waste and recycling trucks that collect household waste at designated times. Even though the street collection system is acknowledged as the most effective method to increase recycling behaviour and attain the highest recycling efficiency, its reliance on fixed schedules may limit convenience for some residents. Consequently, to adopt recycling as a daily routine, there is a pressing need to expand recycling infrastructure to accommodate diverse lifestyles and enhance accessibility.

Moreover, recycling knowledge is also essential to ensure that the public is capable of categorising waste accurately and implementing recycling actions effectively. This study result supported Wang et al.’s research, which suggested that greater awareness of trash classification and recycling will raise the probability of engaging in recycling behaviour [20]. Since 2005, Taiwan has implemented a mandated rubbish classification scheme, and school-based educational programs have assisted in educating Taiwan citizens about recycling. However, proper recycling and garbage classification knowledge need to be disseminated through a broader range of communication channels to ensure that the information is accessible to citizens across all age groups and socio-economic backgrounds.

In reference to TPB, subjective norm, which includes the influence of family, friends, and society, is one of the key factors that influence individual behavioural intention. However, the indicator ‘others influence’ that received the lowest weightage in the overall ranking suggests that Taiwan consumers are comparatively less motivated by social conformity or external expectation in recycling behaviour. This could be due to a cultural shift from traditional collectivist norms towards individualism with the rise of Western values and globalization. Therefore, policies that are grounded on social norms, such as public awareness campaigns or community-based approaches, might be less effective than those based on economic incentives.

Based on the outcome, the recycling behaviour is considered to be gender-neutral, supporting prior studies that revealed recycling, which has been seen as a household activity, is no longer restricted to women as gender equality is promoted [77]. The gender participation rate in recycling is currently fairly balanced, suggesting that Taiwanese waste management continues to progress towards gender equality. While both genders respond positively to rewards, their motivational pathways differ subtly. As such, dual-track outreach strategies can be considered to accommodate the priorities of both genders. For instance, the installation of smart recycling infrastructure that offers recycling information.

Multiple age groups have also resulted in different motivational elements, even though they retained an emphasis on positive reinforcement, recycling knowledge, and basic infrastructure. According to earlier research, the elderly and young adults in their 20s to mid-30s are more likely to participate in recycling activities when they have better access to recycling information and spare time [42,78]. The result showed an opposing conclusion: of all age categories, older citizens over 60 participated in recycling the least. This deviation could be attributed to methodological limitations—particularly the online nature of the survey, which may have reduced older adults’ participation. Nonetheless, the data suggests that older consumers are more likely to be motivated by tangible rewards rather than environmental or educational appeals. While young Taiwanese’s participation in recycling is rather high in this study, the findings also pointed out that improved planning and construction of recycling facilities can further encourage recycling, which is in line with the findings of other studies [35]. Consumers aged 31–40 demonstrated a need for improved recycling knowledge, presumably due to a prolonged absence from formal education on the subject. Consumers over 41 stressed the economic reward that offers tangible incentives and a sense of personal benefits, indicating that as consumers age, their preferences increasingly align with economic and practical considerations. Given the different priorities of consumers from varied age groups, the lifestyle of all target groups should be taken into consideration in policy development to ensure public needs are met.

Studies have shown that workers who recycle often have lower educational levels and that people with higher education levels engage in recycling behaviours at a lower rate [20,25,35]. The recycling behaviour in Taiwan was not significantly determined by education level, although factors affecting their participation varied. As consumers’ education level becomes higher, they place a greater emphasis on the value of recycling knowledge in promoting recycling, presumably as a result of their increased exposure to environmental knowledge and climate change. While recycling education has been integrated into Taiwan’s primary school curriculum, the findings highlight the need to expand educational outreach across all age groups and educational levels, particularly through informal and lifelong learning platforms.

The result showed that consumers of low- and no-income were less likely to recycle, although the majority of them were homemakers, students, or retirees who spent more time at home. The reason behind this preference may be due to the limited access to recycling infrastructure or less exposure to recycling education. According to TPB, low perceived control over the ability to recycle (limited facility) and weaker attitude towards recycling (lack of recycling awareness) reduce behavioural intention. They perceived multimedia communication as the most motivating factor in boosting recycling, consistent with other research that indicated social media subscription accounts are a successful strategy and that publicity in household trash management tends to be propaganda-centred. As consumers’ income increased, their emphasis shifted from incentives to recycling knowledge, possibly due to increased environmental awareness. As seen by the differences in indicators that consumers with different income levels have selected, economic disparity has an impact on consumers’ life quality and their decision to recycle.

Based on the result, family recycling habits can influence individual recycling participation, particularly on the amount of time spent on recycling and attitude towards recycling. This result supported earlier research that stated recycling could evolve into a habitual action if people must engage in it every day and regard it as a civic duty or a basic component of their life routine [44]. Consumers were more likely to remain committed to recycling if they grew up in an environment that adopted these practices, suggesting that family-centred recycling programs could be effective in boosting public recycling participation. Beyond family influence, personal attitude, time consumed, and trust in the recycling system can significantly affect one’s intention to recycle. Policy should also incorporate a targeted educational programme aimed at bridging the knowledge gap, alongside the provision of recycling facilities to encourage the adoption of recycling practices among non-recyclers.

4.1. Recommendations to Improve Recycling Policies

The results align with the TPB, which proposes that consumer behaviour is driven by intention and primarily shaped by PBC in this study. The top three priorities identified imply that enhancing the perceived benefits, convenience, and informational support for recycling can foster public intention to engage in recycling. Accordingly, future recycling policies should be formulated revolving around these 3 key drivers: positive reward, basic infrastructure, and recycling knowledge.

4.1.1. Positive Rewards

The result has demonstrated that positive rewards that offer immediate and tangible benefits are essential for promoting recycling. A rewarding system for recycling is therefore highly recommended, given its success in the European region. Incentives may take the form of monetary rewards, vouchers, or goods.

For example, Germany’s bottle deposit scheme (Pfand System) requires consumers to pay a small deposit on beverage containers, which is refunded upon return via reverse vending machines in supermarkets [79]. This scheme has been implemented statutorily in Germany since 2003 to promote recycling while providing economic benefits, so fostering a circular economy in the nation. Similarly, Hong Kong’s GREEN$ Electronic Participation Incentive Scheme, launched in 2020, uses a smart card or mobile app to award “Green Coins,” redeemable for essential household items at community recycling stations.

While Taiwan already launched a Recycling Rewards Service System in 2011, the system could be enhanced by adapting elements from these international models. Improvement should aim not only to promote recycling, but also alleviate the financial burden on urban dwellers, especially those with low- or middle-class incomes. This will require the establishment of comprehensive policies and regulatory frameworks that both structure the incentive mechanisms and mandate producer participation in reverse logistics and sustainable packaging design. Moreover, the expansion of reverse vending machines and accessible recycling stations is necessary, alongside comprehensive public education campaigns aimed at normalizing return-and-earn behavior. A pilot program could initially be launched in densely populated urban areas, such as Taipei and Kaohsiung, to evaluate feasibility and effectiveness prior to a nationwide rollout.

4.1.2. Basic Infrastructure

Investment and construction of more recycling infrastructure is another critical strategy in increasing consumers’ participation in recycling. Addition and upgrade of recycling infrastructure is the top priority among Taiwan’s younger consumer, possibly due to their high mobility and consumer-driven habits, demanding a convenient and time-saving approach in encouraging recycling. It is recommended that additional recycling bins and collection stations be strategically installed in high-traffic areas, particularly to facilitate users who struggle to adhere to the garbage truck schedule. This includes locations such as public parks, transit hubs, night markets, and commercial streets. Increasing the frequency of waste collection in densely populated areas can help manage waste overflow, improve hygiene, and foster a positive public perception of recycling. Additionally, the expansion of four-in-one 24-h smart recycling machines—currently only available in three heavily populated districts of New Taipei City: Banqiao, Sanchong, and Xinzhuang, should be prioritized [80].

Smart recycling infrastructure integrating innovative technology such as automatic waste sorting, waste compression, and interactive digital screens for user education should be implemented to enhance convenience and simultaneously deliver recycling information. Mobile recycling applications and digital incentive systems can further increase accessibility and participation across all socio-economic groups.

4.1.3. Recycling Knowledge

Recycling knowledge is essential in increasing public participation in recycling, mainly because adequate recycling knowledge is required to conduct accurate waste classification. Being one of the most influential motivators, recycling education should expand beyond the elementary school curriculum and be integrated throughout the education system. This may include project-based learning, science experiments, co-curricular programmes, recycle-themed games, or interschool competitions in nurturing environmental consciousness and positive attitudes towards waste management among students.

International examples offer useful models: a science fair project using recyclable materials practiced in the United States, and Osaki Town in Japan, which conducts global education in elementary and junior schools using recycling business as teaching materials [81,82].

In addition to school-based education, family influence plays a pivotal role in shaping recycling habits. Programmes that promote family-focused recycling activities, such as educational field trips to recycling centres, home D.I.Y. craft using recyclable materials, or composting using bio-waste, can foster high environmental consciousness across generations. To ensure equitable access to recycling information, the government should leverage diverse communication channels. For instance, traditional media like radio and television should target older demographics, while short videos and social media campaigns can effectively engage youth and middle-aged consumers. A multi-channel approach ensures that information reaches all population groups and promotes inclusive participation in recycling practices.

5. Conclusions

In summary, this study examined the key factors influencing consumer participation in recycling in Taiwan through an online survey designed using AHP. The findings identified three primary motivators that significantly enhance public engagement in recycling: implementation of a positive reward system, provision and improvement of recycling infrastructure, and dissemination of adequate recycling knowledge. Considering the minor difference in the weightage of these top three indicators, the findings suggest that they are of nearly equal importance, necessitating a policy framework that holistically incorporates all three key drivers to effectively promote recycling behaviour. In addition, the low weighting score of the ‘other influence’ indicator implies that policies relying on social norms strategies, such as emphasizing majority recycling behavior within the community, may have limited effectiveness in the context of Taiwan. Instead, demographic traits such as age, education level, income, and family recycling habits appear to have a more nuanced impact on individual motivation, shaped by variations in mindset, socioeconomic status, and life priorities. In order to increase public participation in recycling, the recommendations to be translated into recycling policies highlighted a formulation of a rewarding system offering economic benefits, expansion and upgrade of recycling infrastructure to fulfil user needs, and dissemination of recycling information in an inclusive and engaging manner across different demographic segments.

The outcome of this study contributes to the enhancement of the recycling system and the increase of the recycling rate in Taiwan by understanding the most influential factors in consumer recycling behaviour. Recycling policies informed by these drivers could reduce MSW and support sustainable resource management. The result may also serve as a reference for other Asian countries with comparable cultural backgrounds and institutional similarities that seek to improve public involvement in recycling.

However, several limitations should be acknowledged. One of the limitations of this study is the limited number of respondents aged above 60 due to varying levels of digital literacy, resulting in the underrepresentation of this age group. Although the online method allowed broad geographic coverage within a constrained research budget, the final sample size was smaller than initially intended, potentially limiting generalizability. Although the sample size is smaller, this study provides insights into the factors that may influence the recycling behaviour of Taiwanese, which can serve as a foundation for future larger-scale studies in improving the recycling system. In addition, the primary limitation of the AHP is its reliance on the subjective opinions of respondents during the pairwise comparison process. Therefore, a clear definition of each dimension and indicator was provided in the questionnaire to prevent misinterpretation and mitigate potential bias. Another limitation of this study is the absence of a practical significance test of the findings. However, this study still informed the preferences of Taiwan consumers in recycling involvement and underscores the need for further research into more impactful interventions.

As consumer is intricately interconnected with waste generation, upstream design, and logistical systems, future research can also explore the design of reverse supply chains and sustainable packaging solutions that effectively facilitate recycling practices. Moreover, while this study focused on the consumer perspective, additional research is needed to assess the roles of businesses, manufacturers, and institutional stakeholders in advancing Taiwan’s circular economy and enhancing national recycling performance.