In recent years, with the sustained and stable development of China’s economy and society, the attraction of China’s living environment to international immigrants has continued to increase. In 2020, the number of foreign persons living in the mainland reached 850,000, which rose from 600,000 in 2010, an increase of 42% (China National Bureau of Statistics, 2021). At the same time, along with a series of policies aimed at encouraging reasonable population migration, various traditional barriers that restrict population mobility have been broken down. Meanwhile, intensive regional competition has made attracting and retaining young talent an important approach to obtaining development advantages. The urban settlement of young talent has been an important issue in China. For example, in recent years, large- and medium-sized cities, such as Hangzhou, Nanjing, Wuhan, Zhengzhou, Chengdu, and Xi’an, have introduced “New Deals” for young talent, beginning a “talent grabbing” war. Retaining young talent for long-term residencies and, thus, developing cities has become a common concern for city administration and demographic researchers. Yet, young talent endures unstable career developments and a lack of social integration, owing to their short stays in new cities. This results in their increased willingness to move among cities, which makes urban settlement challenging [1
]. The theory of planned behavior (TPB) suggests that psychological factors (such as attitude, subjective norms, and perceptual behavioral control) indirectly influence individual behavior through behavioral intentions [2
]. Based on the TPB, scholars such as Weng and McElroy, proposed the concept of talent urban settlement intentions [3
]; that is, the willingness of talent to work and live in a city for a long time. From this concept we can regard “settlement intentions” as a kind of attitude and behavior tendency, which reflects a comprehensive evaluation of talent from various aspects of a city and is a good indicator of a city’s ability to retain talent. A comprehensive analysis of the influencing factors of the willingness of talent to settle down can help city administration understand what such talent need from the city and thus enact targeted and efficient public policies and provide public goods.
Various studies have been conducted on urban settlement intentions and their influencing factors on population migration between countries, cities, and urban–rural areas. Related studies show that economic factors (e.g., income level and cost of living) [4
], personal factors (e.g., education, occupation, and mobility distance) [5
], amenities (e.g., entertainment and recreational facilities, cultural facilities, consumer shopping, green parks, and public cultural activities) [7
], policy (e.g., household registration and subsidies) [8
], and social factors (e.g., family and friends’ connections, traditional attitudes, and cultural inclusion) [10
] are all important influencing factors in urban settlement intentions. In addition, when examining the influencing factors of overseas immigration, Poprawe believes that political corruption can easily lead people to emigrate abroad [11
]; Bertoli and Moraga found that the migration situation between the two countries is affected, not only by various factors between the two countries, but also by the policies of a third country [12
In recent years, as environmental hazards, e.g., air pollution, have become increasingly serious, more individuals consider air quality an important factor when choosing a place of residence (for physical and mental health purposes). A survey conducted by New Fortune magazine in 2013 found that the environment had become one of the important factors to promote to Chinese residents to consider international migration [13
]. Based on city level data, Qin and Zhu’s research confirmed that, during a period of increasing air pollution, people retrieve “immigrants” through the Internet more frequently [14
]. Kohlhuber et al. found that highly educated people are more concerned about air quality [15
]. Jacquemin et al. further found that educational attainment was highly correlated with the level of perceived annoyance [16
]; particularly, respondents with a graduate degree or higher were found to be the most sensitive to poor air quality. These findings suggest that young talent are more sensitive and concerned about air pollution, and that this may consequently influence their choice of cities when seeking long-term employment. However, existing studies related to population migration settlement have often neglected the topic of air quality [17
]. Further, the impact mechanism of air pollution perception on urban settlement intentions has not been examined, although perception is considered important for initiating actions. With increasing concerns about air pollution, it is necessary to introduce air quality factors into analyses, focusing on their impact on the urban settlement intentions of young talent in China.
This paper is devoted to studying the impact of air pollution perception on the willingness of young talent to settle down in Chinese cities, based on the perspective of the inter-city talent competition using young talent as the research subjects. This study further shows how, and under what circumstances, air pollution perceptions affect the willingness of young talent to settle down in cities. Specifically, this study proposes urban residential satisfaction as a mediating variable and empirically tests its mediating role. It also suggests and tests the moderating role of place attachment in the relationship between air pollution perception and urban settlement intentions. The study systematically investigates the impact mechanisms and boundary conditions of air pollution perception on urban settlement intentions. It further provides theoretical references for city administration to improve the urban talent environment, urban talent attraction, and their retention. The study is structured as follows. Section 2
reviews existing studies and proposes hypotheses for testing. Section 3
introduces the research method and design used in this study. Section 4
presents the empirical results and conducts an in-depth discussion to interpret the results. Section 5
concludes this study by determining future studies.
4.1. Reliability and Validity Tests of the Questionnaire
This study used Cronbach’s alpha coefficient to test the reliability of the measurement’s variables to ensure the reliability and validity of the questionnaire. The results showed that the reliability of the four scales of place attachment, residential satisfaction, air pollution perception, and urban settlement intentions were 0.868, 0.974, 0.912, and 0.934, respectively. All four results were greater than 0.7 and, thus, had good reliability [52
]. Further, according to the method suggested by Fornell et al. [53
], this study used AMOS 24.0 (software to analyze structural equation modeling) to conduct a confirmatory factor analysis (CFA) on the four main variables to calculate the square root of the average variance extracted (AVE) value of each variable. The discriminant validity of each variable was tested by comparing the square root of the AVE value of each variable with the correlation coefficient between the latent variables, as shown in Table 2
. The square root of the AVE value of all the variables was greater than the correlation coefficients. This indicated a good discriminant validity among the variables.
In addition, this study conducted a structural validity test on the four variables. The results showed that all the factor loading values in the four-factor model (model fit indices: X2/df = 4.684, RMSEA = 0.078, IFI = 0.925, CFI = 0.918) (X2 denotes chi-square test, which can assess overall fit and the discrepancy between the sample and fitted covariance matrices. df denotes model degrees of freedom. The chi-square value and model degrees of freedom can be used to calculate a p-value. Model is good fit if p-value > 0.05. RMSEA is an abbreviation for Root mean Square Error of Approximation. It is a parsimony-adjusted index, which is good fit if RMSEA < 0.08. IFI is an abbreviation for incremental fit index with values greater than approximately 0.90. CFI is an abbreviation for comparative fit index, which is good fit if CFI ≥ 0.90) were significantly higher than the general recommendation of 0.4, indicating that the measurement items of each variable could be better aggregated and effectively reflect the same construct. The results also showed that the four-factor model substantially fit indicators better than the other factor models. In summary, the tests described above indicate that the data of this questionnaire have high reliability and validity.
4.2. Testing the Main Effect
Models 1 and 2 in Table 3
show that, after controlling the relevant variables, the independent variable of air pollution perception significantly impacts the dependent variable of urban settlement intentions. The R2
changes significantly, supporting H1 (β = −0.077, p
< 0.01). Particularly, it is worth pointing out that, among the control variables, family and friends in Hangzhou and career development expectations have a significant impact on urban settlement intentions. In other words, migrant talent with family and friends in Hangzhou are more willing to settle in Hangzhou permanently. Additionally, development expectations are also a key influencing variable of urban settlement intentions. Development expectations depend on one’s judgment of future employment and development prospects; the better the expectation, the greater the cost of “giving up.” In 2019, the added value of Hangzhou’s core digital economy industry was CNY 379.5 billion, which increased by 15.1% compared with 2018. Contrastingly, the growth was 14.6% for the e-commerce industry, 13.6% for the Internet-of-Things industry, and 15.7% for the software and information service industry. Such growths resulted from the rapid development of high-tech industries, which provide a career platform for young talent and raises their development expectations. These improvements made Hangzhou one of the top cities in China, in terms of young talent inflow.
4.3. Testing the Mediating Effects
Models 3–6 tested the mediating effects of air pollution perception on residential satisfaction and further influence on the urban settlement intentions of young talent. Firstly, according to the results of Models 3 and 4 in Table 3
, the control variables, such as age and career development expectations, had a significant influence on residential satisfaction and remained robust both in Models 3 and 4. After the inclusion of the key independent variables air pollution perception in Model 4, its effect on residential satisfaction was significant (β = −0.167, p
< 0.001), with R2
changing to 0.024, further enhancing the model’s explanatory power. Therefore, Hypothesis H2a was supported. Further, intermediary variables were included in Model 5. The results showed that residential satisfaction had a significant influence on the urban settlement intentions of young talent (β = 0.530, p
< 0.001). The model’s explanatory power increased by 23.3%, based on the amount of change in R2
; thus, supporting Hypothesis H2b. In addition, this study further examined the effect of perceived air quality and residential satisfaction on settlement intentions. Model 6 incorporated both the independent variable of air pollution perception and the mediating variable of residential satisfaction. The empirical results showed that residential satisfaction had a significant influence on settlement intentions (β = 0.532, p
< 0.001), while the influence of air pollution perception became insignificant (β = −0.012, p
> 0.05). After adding both the mediating and independent variables, the independent variable in the model became insignificant. Contrastingly, the mediating variable remained significant, according to the evaluation method of Baron and Kenny [54
]. This indicates that residential satisfaction plays a mediating role between air pollution perception and urban settlement intentions altogether, thus verifying Hypothesis H2.
To further test the mediating effects, this study conducted a bootstrap test using the PROCESS macro for SPSS/SAS developed by Hayes [55
], and repeated the sample 5000 times. The results showed that the indirect effect of air pollution perception on urban settlement intentions through residential satisfaction was 0.0886, with a 95% confidence interval of [0.055, 0.124] and p
< 0.001. According to the criteria proposed by Preacher and Hayes [56
] for testing mediating effects, if the confidence interval of the indirect effect does not include 0, then the indirect effect reaches a significant level. The empirical results show that the exclusion of a value of 0 also confirmed Hypothesis H2.
4.4. Testing the Moderating Effects
Hypothesis H3a proposes that place attachment has a positive moderating effect on the relationship between air pollution perception and residential satisfaction. In other words, greater place attachment can weaken the effect of air pollution perception on residential satisfaction and offset the decrease in residential satisfaction caused by poor air quality. To test this hypothesis, residential satisfaction was set as the dependent variable per the three-step test method of moderated hierarchical regression analyses. Hierarchical regression was established to sequentially incorporate the control variable, standardized independent variable, moderating variable, the product of the moderating variable, and independent variable into the equation, as shown in Models 7 and 8 in Table 4
. The interaction term insignificantly influenced residential satisfaction (β = 0.004, p
> 0.05). Therefore, Hypothesis H3a was not verified. Further, this study used the same approach to test the moderating role of place attachment between residential satisfaction and urban settlement intentions. Models 9 and 10 showed the empirical results, where the product term had a significant influence on settlement intentions, with β = 0.024 (p
< 0.01) and ΔR2
= 0.132 (p
< 0.05). Therefore, Hypothesis H3b was verified; in other words, the relationship between residential satisfaction and urban settlement intentions is significantly stronger when place attachment is stronger.
To visualize the moderating role played by place attachment in the relationship between residential satisfaction and settlement intentions, this study plotted the moderating relationship based on the method recommended by Aiken and West [57
]. Each chosen variable had one standard deviation above and below the mean. Further, the moderating relationship is plotted in Figure 2
, and shows the difference in the relationship between residential satisfaction and settlement intentions when young talent has different levels of place attachment to the city. With greater place attachment, a slight change in residential satisfaction promotes an increase in urban settlement intentions. Conversely, a weaker place attachment weakens the positive impact of residential satisfaction on urban settlement intentions.
Moreover, this study uses the PROCESS plug-in for SPSS/SAS developed by Hayes to further test the adjusted mediating effect by repeating the sample 5000 times using the bootstrap method. The empirical results show that the mediating effect of residential satisfaction between air pollution perceptions and urban settlement intentions differs significantly at different levels of place attachment. Air pollution perceptions had a stronger indirect negative effect on urban settlement intentions at lower levels of place attachment (β = −0.071, p < 0.001), whereas at higher levels of place attachment, it had a weaker negative effect on settlement intentions (β = −0.038, p < 0.001), (Δβ = −0.033, p < 0 05), with a 95% confidence interval of [−0.021, −0.001], which does not include 0. Therefore, Hypothesis H3 was also empirically supported. With a greater level of place attachment, young talent may still experience residential dissatisfaction due to poor air quality. However, the impact of residential dissatisfaction on urban settlement intentions is weakened by greater place attachment.
Based on data obtained from 987 questionnaires collected from a sample group of young talent in Hangzhou, this study explored the influence of perceived air quality on urban settlement intentions. Hypotheses 1, 2, 2a, 2b, 3 and 3b were supported, whereas Hypothesis 3a was rejected. The findings of this study are as follows. First, air pollution perceptions significantly influence the urban settlement intentions of young talent; the poorer their air pollution perception, the weaker the urban settlement intentions. Second, residential satisfaction significantly mediates the relationship between air pollution perception and urban settlement intentions. In other words, a poorer air quality perception reduces residential satisfaction among young talent, thus weakening their urban settlement intentions. In conclusion, place attachment has a significant moderating effect on the relationship between residential satisfaction and settlement intentions. However, it insignificantly affects the relationship between air pollution perception and residential satisfaction. Compared with a weaker level of place attachment, a greater level of place attachment does not change the dissatisfaction associated with poorer perceived air quality. Instead, it can weaken the effect that this dissatisfaction has on settlement intentions.
Given the current situation of talent competition among various cities, this study focuses on the urban settlement intentions of young talent at an earlier stage. For the first time, it introduces the perceived air quality factor into the analysis of the influencing factors of the urban settlement intentions of young talent. This study examined the influence mechanism of air pollution perception on the urban settlement intentions of young talent. This study also verified the applicability of the demographic characteristics, economic development, and socio-cultural factors proposed by Woon et al. [4
], which influence population migration, on young talent. The idea that young and highly educated people are more concerned about air pollution, as suggested by Jacquemin et al. [16
], was expanded and confirmed, further corroborating the idea that young talent also choose to “vote with their feet” in the face of environmental pollution, as demonstrated by Banzhaf and Walsh [58
]. In addition, unlike in previous studies, which used objective indicators to examine the impact of specific air pollutants on population migration and settlement, this study explored the psychological mechanism of the impact of air quality on the settlement intentions of young talent, starting from the concept of perceived air quality and with the help of factors, such as residential satisfaction and place attachment. This line of exploration is a useful supplement to studies on the impact of objective air quality on the spatial mobility of labor forces. Furthermore, based on the TPB, this study explored, for the first time, the influence mechanism of the relationship between perceived air quality and the settlement intentions of young talent. It was found that residential satisfaction mediates the relationship altogether, thus uncovering the transmission mechanism from better and worse perceived air quality to the strength of urban settlement intentions. The study demonstrated that air pollution perception significantly influences the satisfaction of young talent with urban living, affecting their settlement intentions. Urban air quality should also become an important aspect of the urban talent environment and be incorporated in the government’s public service provisions; thus, the study provides a reference for subsequent studies related to talent environment construction and evaluation. Fourth, this study tested the moderating role of place attachment, an important concept in geography that refers to the emotional interaction between people and specific places, in the relationship among the independent (air pollution perception), mediating (residential satisfaction), and dependent (settlement intention) variables. The findings of this study expand Li and Zhou’s research [45
] on the moderating role of place attachment. Additionally, based on the results of the empirical analysis, it was verified that, in terms of human–place interaction, place attachment does not have a moderating effect on the relationship between air pollution perception and residential satisfaction. Place attachment, however, has a moderating effect on the impact of residential satisfaction on settlement intentions. Therefore, this study explained, in more detail, the influence mechanism of air pollution perception on the urban settlement intentions of young talent.
This study empirically examined the significant influence of air pollution perception on the urban settlement intentions of young talent and explained its influence mechanism. In the context of fierce competition for talent across various cities, this study’s findings serve as an important reference for cities to construct an advantageous ecological environment for their talent and enhance the city’s competitiveness for young talent. First, various cities are currently attracting talent through different policies, such as relaxing household registration restrictions and providing subsidies. However, the influence of environmental factors, such as air pollution on young talent, should not be ignored. Air pollution management should be enhanced through the strategic construction of an urban talent environment to promote high-quality development. Air quality should further be improved continuously through measures such as adjusting industrial structures and strengthening pollution control and dust management in key industries. Second, as improving air pollution control involves complex factors, such as industrial transformation and upgrades and synergy between multiple locations, it must be a long-term process. Greater place attachment can offset the reduction of urban settlement intentions caused by low residential satisfaction due to air pollution. Based on relevant sources, city managers can enhance place attachment to their respective cities by strengthening and improving publicity, creating a compassionate and welcoming image, building community exchange platforms to facilitate social integration, creating opportunities to attract young talent to participate in urban governance, and increasing the understanding and identity of young talent within the city. In conclusion, guided by the various housing needs of young talent, it is necessary to improve the supply of various public goods in the city, enhance urban governance, and implement multiple measures to improve residential satisfaction in the city.
This study, however, has some limitations. First, the survey data were obtained from Hangzhou. Although the inflow of young talent to Hangzhou in recent years has been among the highest in China and the survey sample shows that the sources of talent are also distributed across the country, the survey is inevitably influenced by the geographical characteristics of Hangzhou. Surveys and studies with greater coverage are yet to be conducted to verify whether the research findings can be generalized to the national level. Second, while semi-structured interviews were conducted with young talent, the formal survey was conducted in a relatively short period, making it difficult to properly reflect the dynamic interactions among variables, such as air pollution perception, residential satisfaction, place attachment, and settlement intentions. Follow-up studies can adopt a longitudinal tracking approach to conduct an extensive analysis of the relationship among these variables. This approach would improve the persuasiveness of the research findings. In conclusion, regarding the influence mechanism of air pollution perception on urban settlement intention, whether other variables can be included in the research model should be further explored in future studies.