1. Introduction
As global warming gradually worsens, improving the environment has become an urgent issue for all consumers. Being one of the leading causes of global warming, 80% of greenhouse gases produced by agriculture come from animal husbandry [
1]. If the current diet structure of human beings is maintained until 2050, the emissions of greenhouse gases and the use of water resources, land, and fertilizers are likely to exceed the limit that the Earth can handle [
2]. The greenhouse gases released from food production account for about 20% to 30% of global greenhouse gas emissions [
3], and if meat is completely eliminated from the diet, it will reduce greenhouse gas emissions by about one third [
4].
According to Pimentel and Pimentel [
5], consumers can change their dietary habits and consumption behaviors to achieve the effects of environmental improvement. In this regard, the alternative food production process, plant-based meat alternatives, consumes 46% less energy, produces 90% less greenhouse gas emissions, uses 93% less land and 99.9% less water than the production of beef [
6]. Plant-based meat alternatives are a novel alternative food that meat eaters should consume daily. It can effectively improve the environment as a novel alternative food. If consumers can reduce meat consumption and gradually switch to plant-based meat alternatives substitutes, it could effectively improve global warming.
“Plant-based meat alternatives substitutes” is a reasonably general term; for this study, the term should include analogs that closely approximate whole-muscle animal meat in texture, flavor, and appearance, as well as reconstituted products that mimic processed meat, such as burgers, patties, sausages, and chicken nuggets. Meat substitutes can be classified as plant-based (soy, pea, gluten, etc.), cellular (in vitro or cultured meat), and fermented (fungal protein) since plant-based protein can be used directly to build meat-like substitutes [
7]. As such, plant-based meat alternatives substitutes, substitutes, or substitutes represent a significant sector of this emerging and rapidly growing industry. These non-traditional foods attract investment, research, consumer curiosity, and media attention. Over the past few years, global food scientists’ reinvestment in alternative research has increased research publications on meat alternatives. As “alternatives” to traditional animal-derived foods, they are often promoted as “healthier” (than meat) and sustainable new foods [
8].
The meat alternatives consumers showed the most probable willingness to purchase were found to be plant-based proteins. Plant-based meat alternatives tend to have a wider range of sources than other alternative protein sources [
9]. Those willing to buy alternatives to plant-based meat mainly chose this option because it tended to be more widely available on the market [
10]. However, not all animal-food alternatives are sustainable; some are even ultra-processed [
11]. Furthermore, in addition to technical barriers to mimicking meat texture and flavor, other potential barriers to food safety and nutrition have not yet been adequately addressed [
7,
12]. For these reasons, some problems still need to be overcome, including technological, sensory, nutritional, health, and safety challenges in the development of the alternative meat market [
13].
Currently, the development of plant-based meat alternative products in Taiwan is in its infancy. However, the price of local plant-based meat alternatives products in Taiwan is high, and the marketing and promotion of plant-based meat alternatives are inadequate. Therefore, it has become critical to identify the key factors that can attract consumers to buy alternatives to plant-based meat. However, few studies have been conducted on the consumer behavior of plant-based meat alternatives, a new type of vegetarian meat. Additionally, the limited research conducted on green foods has tended to use the Theory of Planned Behavior for research and discussion, but little research has been conducted on evaluating consumers’ perception of green value by their perceptions of its green features and price. Finally, research on whether consumers’ perceived animal welfare value affects their consumption attitudes and purchase behaviors towards plant-based meat alternatives is also limited. This study aimed to conduct a behavioral study on consumer values and attitudes toward plant-based meat products in Taiwan.
Based on the value-attitude-behavior (VAB) model, this study deconstructs the characteristics of the key factors that influence consumers’ purchase behavior. Furthermore, Torri et al. [
14] noted that younger consumers are more receptive to new foods. Therefore, this study was designed to investigate the key factors that influence the purchasing behavior of plant-based meat alternatives using the VAB model with college students in Taiwan as primary research subjects. Secondly, it investigated whether the novelty of such plant-based meat alternative products can act as an intervening variable to affect consumers’ subjective judgment of new food products. Finally, the relationship between product knowledge and behavior was explored regarding the amount of knowledge they had about the product. As a result of the analysis, this study not only deconstructed the key factors that influence consumers’ attitudes and purchase intentions toward plant-based meat alternatives but also suggested marketing strategies for the future introduction of new products such as plant-based meat alternatives. This study is expected to provide suggestions for the development of alternative food industries, which will lead to a change in dietary patterns to achieve the ultimate goal of environmental protection.
4. Results
4.1. Reliability and Validity
The primary function of the reliability analysis is to test whether the results of each variable measurement are stable and consistent, and to what extent. If the value of the reliability coefficient is higher, the internal reliability of this measurement is consistent and reliable [
99]. The Cronbach’s alpha value should be at least greater than 0.50, preferably greater than 0.70 [
91]. The Cronbach’s α of each construct in this study was more significant than 0.7 (see
Table 3 for details), indicating that the measurement tool had considerable reliability. That is, all the measurement elements of the six constructs of the scale had internal consistency, and a high degree of stability of the questionnaire was maintained. The higher the value of composite reliability (CR), the higher the proportion of the actual variance to total variance, that is, the higher the internal consistency, which can be considered as the internal consistency of a construct. The composite reliability value (CR) should be higher than 0.6 [
100]. The CR of the variables in this study ranged from 0.781 to 0.912, indicating that this model had good internal consistency. The extracted average variance (AVE) is the degree to which all the variables measured in the latent variables can explain the latent variables. The higher the AVE, the higher the degree of the latent variables that are explained by the measured variables. The AVE for each factor was between 0.487 and 0.784, which is higher than the recommended reference of 0.36 [
101]. However, the factor loading value (0.53–0.92) was higher than the recommended level of 0.5 [
102]. Means, standard deviations, and correlations among the constructs are presented in
Table 4.
Both perception of green value and animal welfare value had a significant positive correlation with attitude (γ = 0.591, γ = 0.458, p < 0.01). That is, when consumers thought that plant-based meat alternatives had a higher perception of green value or animal welfare value, their attitude towards plant-based meat alternatives also became more positive. There was also a significant positive correlation between attitude and behavior (γ = 0.699, p < 0.01), that is, when consumers had more positive attitudes towards plant-based meat alternatives, the purchase behavior of plant meat also increased. There was a significant positive correlation between product knowledge and behavior (γ = 0.571, p < 0.01). In other words, when consumers believed they had more product knowledge about plant-based meat alternatives, their purchasing behavior for plant-based meat alternatives also increased. There was a significant positive correlation between the novelty of plant-based meat alternatives products and the perception of green value and animal welfare value (γ = 0.440, γ = 0.301, p < 0.01). If consumers were more receptive to the novelty of plant-based meat alternative food, their perception of green value and animal welfare value increased too. There was a significant positive correlation between the novelty of plant-based meat alternatives and attitude (γ = 0.471, p < 0.01). This means that if consumers were more receptive to the novelty of the food of plant-based meat alternatives, their attitudes toward it would be more positive. There was a significant positive correlation between the novelty of plant-based meat alternatives food and behavior (γ = 0.580, p < 0.01). If consumers were more receptive to the novelty of plant-based meat alternatives products, the purchase behavior of plant-based meat alternatives would also increase.
4.2. Structural Equation Modeling and Empirical Analysis
This research applied SEM using AMOS 22.0 to assess the path relationships among the perception of green value, the value of animal welfare, attitude, purchase intention, and knowledge of the product. The results indicated that the measurement model provided a good fit for the data (χ
2/df = 2.351, GFI = 0.930, AGFI = 0.900, CFI = 0.964, NFI = 0.940, SRMR = 0.049, RMSEA = 0.060). The χ
2/df ratio is below the value of 3 (Carmines, 1981), the GFI, AGFI, CFI NFI, and IFI exceeded the recommended threshold of 0.90 [
96,
97,
98,
99,
100,
101,
102,
103,
104], and the values of RMSEA and SRMR were below the cutoff value of 0.08 [
105]. This indicated that the approach used in this study to model the examined data was appropriate. The hypotheses that test the results of the model data are provided in
Figure 2 and
Table 5.
The path coefficient of perception of green value and attitude has reached a significant level (γ11 = 0.769, p < 0.001), indicating that perception of green value and attitude were positively and significantly correlated, meaning that the higher the consumers’ perception of green value, the higher their attitude towards plant-based meat alternatives. So, H1 was supported. The path coefficient between animal welfare value and attitude reached a significant level (γ12 = 0.246, p < 0.001), indicating that animal welfare value and attitude were positively and significantly correlated, i.e., the higher the consumers’ animal welfare value of consumers, the higher their attitude towards plant-based meat alternatives. Therefore, H2 was supported by empirical data. The path coefficient of attitude and behavior reached a significant level (β12 = 0.719, p < 0.001), indicating a positive and significant correlation between attitude and behavior, which means that the higher the attitude towards plant-based meat alternatives, the higher the purchase behavior of plant-based meat alternatives. Therefore, H3 was supported by empirical data. The path coefficient of knowledge and behavior reached a significant level (γ23 = 0.450, p < 0.001), indicating that knowledge and behavior were positively and significantly correlated, showing that the more knowledge consumers have about plant-based meat alternatives products, the higher the purchase behavior of plant-based meat alternatives. Therefore, H4 was supported.
4.3. Testing Interference Effects
Hierarchical regression was used to analyze the interference effect of the novelty of plant-based meat alternatives on the relationship between knowledge of the product and purchase intention. The analysis results are presented in
Table 6. The regression model M1 in this table shows the impact of knowledge of the product on purchase intention in the first stage, with the value β = 0.571 and
p < 0.001. The coefficient reached a significant level and presented a significant positive correlation. The regression model M2 shows the influence of the novelty of plant-based meat alternatives on the intention to purchase in the second stage, with its value β = 0.393 and
p < 0.001. The coefficient reached a significant level and showed a positive and significant correlation. The regression model M3 shows that after adding the product term of product knowledge x novelty of plant-based meat alternatives in the third stage, the variance explained (R
2) did not change, the change value of the variance explained (ΔR
2) was 0.000, and its value was β = 0.014,
p > 0.05. The coefficient did not reach a significant level, indicating that the novelty of plant-based meat did not interfere with the relationship between “product knowledge and behavior”. From the regression model M3 in
Table 6, it can be seen that by adding “if the average number of the novelty of plant-based meat alternatives products was taken as the cut-off point”, the samples were divided into two groups, with high-novelty of plant-based meat alternatives food and low-novelty of plant-based meat alternatives food for separate analysis; the line segment of the regression model of product knowledge and purchase intention drawn was as shown in
Figure 3. Whether consumers’ preference for the novelty of plant-based meat alternatives was high or low, the effect between product knowledge and purchase intention could not be increased. Therefore, Hypothesis H5 was not supported.
6. Contributions and Limitations
6.1. Contributions
(1) Research Using VAB Model
Although the theory of planned behavior (TPB) has been used extensively to study consumer behavior in the past, this study chose to adopt the VAB model, which can be used to investigate the characteristics of different products and services through different value variables. The results of this study showed that this model could be used to investigate the value, attitudes, and behaviors of consumers toward green foods such as plant-based meat alternatives.
(2) Research on Perception of Green Value and Animal Welfare Value
There are many studies on green perceived value and animal welfare value, but there are few studies on novelty food using these two values simultaneously. This study collated the unique values of plant-based meat alternatives products and further explored them using the VAB model. It was found that both values could positively and significantly affect consumer attitudes toward plant-based meat alternatives, and the findings of this study will be helpful for future studies by future researchers.
(3) Research with Novelty of Plant-Based Meat Alternatives as an Intervening Variable
There have been many studies on the novelty of foods in the past, but very few studies have used it as an intervening variable. In this study, the novelty of plant-based meat alternatives was used as an intervening variable, and it was found that the relationship between knowledge of the product and behavior was not influenced by the moderation of the novelty of plant-based meat alternatives. This means that the relationship between product knowledge and behavior was not affected regardless of the high or low novelty of the food. This research finding will help future researchers in their research development.
6.2. Limitations
Regarding research limitations, first of all, this study faced the limitation of sampling the research participants. For Taiwanese consumers, the novelty of plant-based meat alternatives is not familiar to university students, so this study tried to choose the department of food science or nutrition of Taiwan universities as the research sample. According to data from the Taiwan Ministry of Education for the 2022–2022 school year, the ratio of male to female students in food science or nutrition is 25–28% male and 72–75% female, which is similar to the percentage of 27.4% to 72.6% in the research sample of this study. Only students in the Department of Food Science or nutrition at Taiwan University were taken as research participants, resulting in a limited sampling scope so that the results of this study cannot be extrapolated to consumers of all age groups. Additionally, the participants have backgrounds in related fields such as food science, environmental science, and nutrition. Your experience will influence their different thoughts or acceptance of your intention to buy plant-based meat alternatives and the study results. Secondly, the study did not analyze demographic attributes (such as age and the amount of disposable monthly). Plant-based meat alternatives currently on the market are novel foods that are not yet popular and are expensive. Therefore, age and the monthly disposable amount will also affect consumers’ purchase intention. Lastly, since the analysis process of this study only focused on the impact of perception of green value and animal welfare on the subjects’ attitudes and behaviors of the explained subjects, and the cumulative variance was 83.562%, it is speculated that plant-based meat alternatives products may have other values (such as environmental protection value, moral value, and various variables perceived value) which have not been explored.
Based on these limitations, this study suggested that future researchers should expand the research subjects to include all age groups or monthly disposable amount with experience in purchasing plant-based meat alternatives, incorporating variables such as moral value environmental value, and experiential value into the exploration, and conduct comparative measurements between age groups and monthly disposable amount through post hoc tests to find the actual main customer groups of plant-based meat alternatives and make the research results more complete.