In the 21st century, scientists, educators, and industry insiders have regarded creativity as one of the skills most critical for personal and organizational development (Henriksen et al., 2019; Pink, 2006; Van Harpen & Sriraman, 2013). Creativity is usually defined as the generation of novel and practical ideas and products (Oldham & Cummings, 1996). More and more researchers have begun to focus on the influencing factors of creativity to enhance individual creativity and national innovation capability. Specifically, scholars have mainly explored the factors of innovation behavior in terms of three aspects: individual characteristics, external environment, and the interaction between individual and environment (Sun, 2018). For example, Yun and Wen (2016) found that proactive personality can promote employees’ innovative behavior. West and Farr (1990) verified the positive effect of team innovation climate on employee creativity. Gu et al. (2014) introduced individual creativity efficacy and positive emotion as mediating variables when discussing the impact of organizational support on employee innovation behavior. Their research showed that organizational support both directly and positively affects innovation behavior, and also indirectly affects innovation behavior through individual creativity efficacy and positive emotion.

Innovation comes from a certain cultural atmosphere, so that cultural and social values directly affect the form of innovation behavior (Sui et al., 2012). Arieti (1976) believed that cultural environment is a kind of creative gene, and stated that individuals’ creativity is the combination of their own potential creativity and the creative gene of their culture. Niu and Sternberg (2001) also believed that the cultural environment in which individuals live has an important impact on the level of their creativity. For example, people have different emphases on the novelty and practicality of creativity according to their culture. In Western culture more attention is paid to the novelty dimension of creativity, whereas in Eastern culture more attention is paid to the practicality dimension (Lubart, 1999). In addition, some cultures emphasize obedience to social norms, avoidance of uncertainty, and respect for authority, which may inhibit the emergence and development of originality (Erez & Nouri, 2010). For example, collectivistic cultures in which obedience to social norms and respect for authority are emphasized are not conducive to the emergence of creative ideas (Erez & Nouri, 2010), and cultures with high power distance hinder creativity and innovation (Efrat, 2014). Other cultures emphasize individualism, low power distance, and acceptance of uncertainty, which are cultural characteristics that have been found to be beneficial to the stimulation of creativity (G. Zhang & Jin, 2014). Thus, we set out to establish the impact on creativity of Chinese traditional culture, especially the impact of Zhongyong thinking, which is generally regarded as the most representative style of thinking in Confucian culture.

Within the cultural setting of China, scholars have written a series of discussions on the relationship between Zhongyong thinking and creativity. For example, Liao and Dong (2015) found that Zhongyong thinking is conducive to the sharing and circulation of knowledge, thus increasing the emergence of new ideas, whereas J. Yang et al. (2012) stated that the risk of offense in innovation behavior is contrary to the harmony and balance advocated in Zhongyong thinking, such that this inhibits the emergence of innovative behavior. In addition, J. Wei (2019) put forward the concept of an inverted U-shaped relationship between Zhongyong thinking and creativity, namely, that when Zhongyong thinking is at an appropriate level, it is most conducive to the development of creativity. Thus, it can be seen that there is still no agreement among scholars on the relationship between Zhongyong thinking and creativity.

Therefore, there is a need to investigate the relationship between Zhongyong thinking and creativity, and also to explore what moderating variables may affect this. Meta-analysis can effectively avoid the deviation of results caused by sample bias or statistical error in a single study so that more accurate results can be obtained. Therefore, we conducted a meta-analysis to examine the effect sizes of the relationship between Zhongyong thinking and creativity and explored possible moderators that may have led to heterogeneity in the studies included.

Zhongyong Thinking

As the central theme in Confucianism, Zhongyong thinking is distinctive and important philosophically in Chinese traditional culture, and has a profound influence on Chinese people’s way of thinking. Pang (1980, 2000) pointed out that Zhongyong is not only a part of a type of supreme morality, but is also how Chinese people understand and interact with the world. Neutralization is its goal, namely, not going to extremes, pursuing harmony, and avoiding leaning to either side (D. Zhang et al., 2001).

Through a review of Confucian classics, C. F. Yang and Chiu (1997) first constructed Zhongyong thinking as a practical thinking system of metacognition from the perspective of real-life practice, including four levels: philosophy of life, individual event handling, reflection after the event, and mental health. After the construction of a Zhongyong practical thinking system, they first introduced it to psychological and management research and conducted many empirical studies to explore its application in these fields. The Zhongyong practical thinking system is constantly being revised and improved and plays an important role in leading the study of Zhongyong in the context of Chinese psychological research (Q. Wei & Han, 2022). On this basis, researchers have put forward their understanding of Zhongyong thinking according to different emphases (Z. Zhou et al., 2019). Chiu (2000) expanded the theory of Zhongyong thinking from the perspective of specific event handling, defining it as a directional mode of thinking that guides individuals to differentiate and analyze opposing elements and forces in the environment, and find out a method of coordination so that these contradictory elements and forces can promote each other, coexist, and grow together, and finally achieve a harmonious and balanced state.

C. H. Wu and Lin (2005) defined Zhongyong thinking from the perspective of integrating different opinions. In line with this view, Zhongyong thinking refers to “consider[ing] a problem from multiple perspectives and mak[ing] a behavioral decision that considers both the self and the whole situation after taking into account different viewpoints in detail” (C. H. Wu & Lin, 2005, p. 255). It contains three dimensions: integration, multithinking, and harmony.

Du and Yao (2015) expanded the structural connotation of Zhongyong thinking from the perspective of Chinese people’s perception and application of Zhongyong thinking in real life. They advocate adopting the way of the mean and congruence to achieve a balance between the individual and the environment. Of these two concepts, mean represents that when there is potential conflict with the environment, individuals should choose moderate and nonextreme behavior, and congruence means that when there is no conflict with the environment, individuals still need to balance the surrounding situation thoroughly and then choose behavior consistent with the external environment.

In addition to these concepts based on the Zhongyong practical thinking system, there are several other concepts related to Zhongyong thinking. Among them, holistic thinking is most often discussed. It is a typical thinking style of Chinese people, which means regarding the world as a complex whole, paying attention to the whole and the relationships between the various parts of the whole, considering problems from multiple perspectives, containing and integrating different views, and establishing new connections (Hou et al., 2016; Morris & Peng, 1994). This is consistent with the connotation of Zhongyong thinking. Researchers have conducted a series of experiments using attention span and other related variables to measure the integrity of holistic thinking. Wang et al. (2013) examined the relationship between Zhongyong thinking tendency and behaviors when viewing banner advertisements. Eye-tracking data suggested that participants with high (vs. low) Zhongyong thinking viewed advertisements of lower complexity through a larger and more scattered scanning path, indicating that they adopted a more holistic strategy to integrate information. Chang and Yang (2014) used a dual-target detection task to further show that participants with a high level of Zhongyong thinking are good at adopting holistic strategies in cognitive processing. Taiwanese participants with a high (vs. low) level of Zhongyong thinking preferred the holistic reaction (Huang et al., 2014). Therefore, holistic thinking is considered to form the core of Zhongyong thinking (Chiu, 2000; Huang et al., 2014).

Researchers have developed different scales to evaluate Zhongyong thinking based on their own viewpoint. C. F. Yang and Chiu (1997) constructed the Zhongyong Practical Thinking Scale, composed of eight facets and 16 forced selections (e.g., “Please always focus on maintaining the overall situation and don’t just think about yourself” paired with “If you always have to take the overall situation into account, you often just compromise”). There are two response options for each item, one in line with and the other against Zhongyong thinking, and respondents are asked to choose one option and express their agreement with the sentence on a 7-point Likert scale ranging from 1 = strongly disagree to 7 = strongly agree. If the selected item violates Zhongyong thinking, the score will be reversed.

Most later scales have used as their base C. F. Yang and Chiu’s (1997) Zhongyong Practical Thinking Scale, then reduced or refined it (see, e.g., Q. Wei & Han, 2022). C. H. Wu and Lin (2005) developed the Zhongyong Thinking Scale with a focus on the process of individual interpersonal interaction, including 13 items divided across the three dimensions of multithinking, integration, and harmony (e.g., “I am used to thinking about the same thing from multiple perspectives,” “When discussing opinions, I will take into account the conflicting opinions,” “When I make a decision, I usually check the harmony of the overall atmosphere”). Respondents rate the items using a 7-point Likert scale ranging from 1 = strongly disagree to 7 = strongly agree, where the higher the score is, the higher is the degree of Zhongyong thinking. Du and Yao (2015) developed a Zhongyong thinking scale consisting of four dimensions: mean and congruence, personal cultivation, no ambition, and passive avoidance. They considered mean and congruence as the core of Zhongyong thinking (e.g., “Be impartial and choose a moderate plan” and “Consider the ideas and practices of people around you”).

To sum up, although there are diverse understandings of Zhongyong, we focused on the practical thinking level involving the way that people approach and deal with situations in everyday life—their tendency in treating and thinking about problems. This is a typical indigenous psychological concept that has been increasingly recognized by researchers in China and cannot be reduced to an existing concept though containing a lot of information. Different understandings and measurements of this concept are all based on the same practical thinking system. Thus, we did not incorporate the diverse conceptualizations based on the theory of Zhongyong thinking in regard to morality and values in this study of the relationship between Zhongyong thinking and creativity. Our focus is on Zhongyong thinking as it is applied in daily living for dealing with everyday problems.

Creativity

Convergent thinking is defined as the ability to select a correct response from a set of alternatives or to deduce this correct response from a set of stimuli. The emphasis is on speed, accuracy, and logic. Therefore, it can be ensured that the ideas are practical (Cropley, 2006) by adopting convergent thinking, respecting rules, and focusing on details (Erez & Nouri, 2010). To generate novel ideas, individuals need to adopt divergent thinking and build new connections between different elements (Von Fange, 1955).

In terms of the method of measuring creativity, some studies measured individual creativity by asking others (superiors, colleagues, or teachers) to evaluate an individual and others asked individuals to self-evaluate their creativity. Researchers have used various measures to assess creativity. At present, the more widely used creativity scales include the three-item Employee Innovation Behavior Scale, in which the supervisor evaluates the creativity of employees (Oldham & Cummings, 1996); a nine-item scale integrating the generation, promotion, and realization of ideas as three stages of employee innovation into a single-dimensional employee innovation behavior scale (Janssen, 2000); the Innovation Behavior Scale, which is a measure of innovative ideas and executive ability (Scott & Bruce, 1994); the Chinese Employee Creativity Scale, which measures professional knowledge of creativity, creative thinking, and internal motivation (J. Zhou & George, 2001); and the measure used by Kleysen and Street (2001), who summarized individual innovative behavior in five dimensions, including opportunity seeking, conception generation, concept evaluation, concept support, and concept application.

Zhongyong Thinking and Creativity

Most researchers believe that Zhongyong thinking can promote individual or team creativity and that the level of Zhongyong thinking is positively correlated with the level of creativity (e.g., Chen et al., 2018; S. Wu et al., 2020; Y. Yao & Fan, 2014; H. Zhou et al., 2017). First, in the definition of creativity, the two dimensions of novelty and practicality are contradictory (G. Zhang & Gu, 2015). Specifically, novelty requires divergent thinking and practicality requires convergent thinking (Van Fange, 1955). Therefore, in creativity, there must be the right balance of both novelty and practicality rather than one or the other predominating (G. Zhang & Gu, 2015). Additionally, the individual needs to find relational creativity between existing elements and cut off the connection with existing elements (IJzerman et al., 2014). It is precisely because of the contradictions inherent in creativity that Zhongyong thinking can help individuals solve these contradictions. Second, the aim in Zhongyong thinking is to maintain balance and harmony, which may create a relaxing atmosphere, resulting in promoting deep communication and friendly cooperation among members of an organization, so as to stimulate creativity (Burke et al., 2006; Martins & Terblanche, 2003). Third, people with a high (vs. low) degree of Zhongyong thinking are more likely to feel positive emotions (Lee et al., 2013), thus inspiring new and exploratory thoughts and behaviors, according to the broaden-and-build process of positive emotion (Fredrickson & Joiner, 2018).

Conversely, some researchers have found that Zhongyong thinking may inhibit creativity. Y. Wu (2009) and X. Yao et al. (2010) showed that Zhongyong thinking emphasizes compromise and tolerance, which obliterates people’s pursuit of individuality and makes them stick to the rules, thus reducing their creativity. However, these studies had a one-sided understanding of Zhongyong thinking as a factor opposite to change, which ignores its deep-seated meaning (G. Zhang & Gu, 2015). For example, if neutralization is misinterpreted as eclecticism and following the herd, this incomplete understanding means that lack of creativity will be attributed to Zhongyong thinking. Thus, in view of these two divergent views on Zhongyong thinking and creativity, the conclusion remains to be explored.

Aim of the Present Study

Several studies have been conducted to investigate the relationship between Zhongyong thinking and creativity. In each of these studies (see Table 1) there were differences in number of participants, measurement tools used, way of measuring Zhongyong thinking and creativity, publication status (published or unpublished), journal type (Chinese domestic or English international journal), gender distribution in the sample, and correlation between Zhongyong thinking and creativity.

Therefore, our aim was to achieve the following three goals through this meta-analysis: First, examine the size of the effect of the association between Zhongyong thinking and creativity; second, to examine whether there are moderators that affect the effect size; and third, to establish whether there was publication bias in these studies. Publication bias refers to the representativeness of the samples in the studies included in the meta-analysis. For example, studies in which the results show a level of significance are more likely to be published, whereas when studies have nonsignificant results the researchers may have difficulty in getting their work published, even though their research contains important information (Rothstein et al., 2006).

Method

Literature Search

Through our literature search we found that most of the empirical research on Zhongyong thinking began in 2009. Therefore, we selected Chinese and English literature from 2009 onward as the research object. We searched the China National Knowledge Infrastructure Journal Full-Text Database, the China Doctoral Dissertation Full-Text Database, the China Excellent Master’s Full-Text Database, Google Scholar, Springer Link, Elsevier, and other databases to collect Chinese and English literature with the key words of Zhong yong, Zhong-yong, Zhongyong, mean, doctrine of mean, creativity, creative, and innovation. Combinations of two-word sets were used for initial keyword filtering (e.g., Zhong-Yong and creativity; Zhongyong and innovation). In addition, we searched the reference lists of every article for more relevant literature. Figure 1 shows the search process and the number of studies located.

Inclusion and Exclusion Criteria

To be included in the final analysis, studies must have met the following criteria:

(a) Studies must be empirical research in which numerical results were reported. Review articles and meta-analysis studies were excluded.

(b) Studies must have used scales measuring both Zhongyong thinking and creativity. Sufficient statistical information to calculate the correlation coefficient between Zhongyong thinking and creativity must have been reported in the literature.

(c) If more than two papers were published with the same data or in different forms, only one of them was included.

(d) Studies must have measured the same person’s Zhongyong thinking and creativity. For example, a study in which the leaders’ Zhongyong thinking and their staff’s creativity were measured was excluded in our meta-analysis (e.g., Hu, 2017).

(e) An exact sample size must be recorded in the study.

Two raters coded the information included in the meta-analysis independently: (a) number of participants, (b) the effect size of the association between Zhongyong thinking and creativity, (c) measurement instruments of Zhongyong thinking and creativity, (d) whether creativity was measured through self-evaluation or evaluation by others, (e) publication status, (f) whether publication was in a Chinese domestic or an international (English) journal, and (g) gender distribution of the sample. The interrater reliability was 91.3%. Ratings that had not yet reached preliminary agreement were discussed later in order to reach a final agreement.

Data Extraction and Conversion

Our aim was to investigate the relationship between Zhongyong thinking and creativity by collecting the correlation coefficients included in the selected studies. If the results included only the correlation coefficient of each subdimension of Zhongyong thinking and creativity, we used the Fisher Z-r conversion table to convert them into Fisher Z scores, calculate their average scores, and then convert them into correlation coefficients for input. In this way we could use the correlation coefficient (r) as the effect index and convert the value of the correlation coefficient in each study into the corresponding Fisher Z score, Then, the Z value was weighted into the correlation coefficient to get the total effect value. In addition, some studies used regression or structural equation modeling, so the correlation coefficient was not reported, but the information we needed to calculate the correlation coefficient was reported. For instance, the value of beta was transformed into a correlation coefficient in the structural equation modeling studies by Chen et al. (2018) and Pian (2019). We used Psychometrica software to estimate the Pearson product-moment correlation coefficient in these studies (Peterson & Brown, 2005) and Comprehensive Meta-Analysis (CMA, Version 3.3) statistical software to compute Fisher Z scores, Pearson product-moment correlation coefficients, and other statistical values in this meta-analysis (Borenstein et al., 2014). Cohen (1988) considered that an effect value ≤ .10 was small, a value between .10 and .40 was medium, and a value ≥ .40 was large.

Data Analysis

We used CMA (Version 3.3) to calculate the overall effect size for the association between Zhongyong thinking and creativity. We used a mixed-effects model to evaluate the measurement method of creativity, the measurement tool of Zhongyong thinking, and the publication status of the article as the three categorical moderators. The measurement method of creativity was coded and divided into two categories: self-evaluation of creativity, comprising studies in which creativity was measured through self-report, and others’ evaluation of creativity, comprising studies in which creativity was measured with evaluation by people other than the respondents. The measurement tool of Zhongyong thinking was differentiated into three categories: (a) The scale developed by Taiwanese scholars C. H. Wu and Lin (2005) in the Chinese language, (b) the scale developed by scholars Du and Yao (2014) in the Chinese language, and (c) other studies measuring Zhongyong thinking that did not fit into either of these two categories. The publication status was coded and divided into two categories: published and unpublished.

Heterogeneity

The Q-statistic was calculated to determine if there was significant heterogeneity present among the studies. For the studies included in the meta-analysis, we found a significant Q-statistic (Q = 353.38, p < .001) and a high I² index (I² = 94.91). Therefore, we believed that the random effects model was the most suitable for our meta-analysis and we performed moderator analyses to account for the variability.

Results

The Overall Association Between Zhongyong Thinking and Creativity

For these 18 studies with 19 effect sizes, we found an overall effect size of r = .37, N = 6,494, 95% confidence interval (CI) [0.27, 0.46], which indicates a medium-strength, positive, and significant relationship between Zhongyong thinking and creativity. Figure 2 shows the forest plot.

Moderator Analyses

The measurement method of creativity (self-evaluation or others’ evaluation) was a significant moderator of the relationship between Zhongyong thinking and creativity (Q = 6.87, p < .01). Specifically, there was a stronger effect size for studies with self-evaluation of creativity (k = 13, r = .44, p < .001) than for studies with others’ evaluation of creativity (k = 6, r = .19, p < .05). The moderating effects of the measurement tool of Zhongyong thinking and the publication status did not reach significance in explaining variability across studies: Zhongyong thinking measurement tool (Q = 2.69, p > .05); publication status (Q = 3.47, p > .05).

Publication Bias

For our meta-analysis we not only collected unpublished research data as much as possible, but also tested the publication bias using two methods: a funnel plot and the test devised by Egger et al. (1997). Funnel plots are essentially scatter charts. If there is no bias, the points in the plot should converge into a roughly symmetrical (inverted) funnel (Rothstein et al., 2006). In the funnel plot (see Figure 3), the research literature is basically distributed on both sides of the average effect, which indicated that the effect of publication bias may be nonsignificant in the studies included in our meta-analysis.

To a large extent, the interpretation of a funnel plot is subjective and rough. Therefore, we also conducted a regression intercept test (Egger et al., 1997). Generally, the intercept and 95% CI of the linear regression equation need to be calculated, and then hypothesis testing is conducted if the intercept is 0. If the result is not significant, it can be considered that there is no publication bias (Egger et al., 1997). The linear regression result was not significant, t(17) = 0.26, p >.05, which indicated that no significant publication bias was found. Thus, based on the funnel plot and the regression intercept test result, there was no significant publication bias in the studies included in our meta-analysis.

Discussion

 

Overall, consolidation of the results showed that across the studies included in our meta-analysis Zhongyong thinking was significantly associated with creativity. The overall effect size was consistent with the results of previous studies (Liao & Dong, 2015; J. Wei, 2019; H. Zhou et al., 2017) and suggests that Zhongyong thinking is closely related to creativity. The reasons for this may be, first, as described earlier, that there is a contradiction between the novelty and practicality of creativity, and the integration and multithinking emphasized in Zhongyong thinking can make the individual deal correctly with contradictions and paradoxes (Chiu, 2000). According to G. Zhang and Gu (2015), Zhongyong thinking can help individuals solve the contradiction between the novelty and practicality of creativity, which is conducive to the proposal of creative programs.

Second, from the perspective of the definition of Zhongyong thinking, neutralization is emphasized as the goal, along with attention to harmony, treating the problem as a whole, considering it from various aspects, and finally choosing a mode of behavior after taking into account the overall situation and the self (C. H. Wu & Lin, 2005; S. Wu et al., 2020). Individuals with a high degree of Zhongyong thinking can maintain harmony with the people around them and the surrounding environment. This harmonious atmosphere can make all members of an organization more likely to exchange information, which is conducive to the development of creativity (Liao & Dong, 2015). Integrity is helpful for individuals in searching for information from a wider space and solving problems creatively (Du & Duan, 2017). Multithinking enables individuals to fully consider problems from multiple perspectives and to observe and discover neglected breakthrough points, thus forming the starting point of innovation activities (J. Wei, 2019). As previous studies have shown, Zhongyong thinking emphasizes inclusiveness, which enables individuals to absorb all kinds of views is conducive to the generation of creative ideas (S. Wu et al., 2020).

Third, in order to conceive new ideas, the previous rules often need to be broken. However, breaking the rules means that there will be conflict and certain risks, which may bring a negative emotional experience to individuals (Kurtzberg & Mueller, 2005). Studies have shown that individuals with Zhongyong thinking pay great attention to harmony with the people around them and the surrounding environment, which can alleviate conflict and reduce the generation of negative emotions (S. Wu et al., 2020). At the same time, this attention to harmony can also lead to forming good interpersonal relationships, increasing positive emotions, and improving individual life satisfaction and happiness (C. H. Wu, 2006; Z. Yang et al., 2014). Furthermore, the constructive function of positive emotions is conducive to expanding cognitive scope and constructing psychological resources (Fredrickson & Joiner, 2018), which promote individual creativity (Schutte & Malouff, 2020).

Our meta-analysis showed that the association between Zhongyong thinking and self-evaluated creativity was significant, and that between Zhongyong thinking and creativity evaluated by others was also significant but to a lower degree. It seems that the reliability of our research results has been weakened because evaluation by others is usually regarded as more objective than self-evaluation. However, there may be several possible reasons to explain this difference.

First, there may be same-source bias owing to the use of the self-report method for assessment of different concepts. In the future, researchers could choose different evaluation methods to avoid common-source inflation of associations. Second, compared with the 13 studies in the meta-analysis in which creativity was self-evaluated, the number of studies in which others evaluated creativity (n = 6) is relatively small. Therefore, these results may be accidental and unstable, even though evaluation by others is more likely to be objective. Third, the studies in which creativity was evaluated by others were mainly based on the evaluation of superiors. In an organizational environment with high Zhongyong thinking, creativity will indeed flourish but, although Zhongyong thinking is conducive to the development of creativity, in the organizational setting of paternalistic leadership in China, leaders mostly suppress the ability of subordinates to establish their authority (Cheng et al., 2000; Lu et al., 2022; Sposato, 2021). In addition, a low evaluation of subordinates’ creativity can help to compensate for subordinates’ shortcomings and make for continued progress, which is also in line with the characteristics of Chinese self-enhancement (Heine et al., 1999). Therefore, the creativity of subordinates may be underestimated when evaluated by superiors. Finally, in studies exploring the influence of leaders’ Zhongyong thinking on employees’ creativity it has been found that Zhongyong thinking plays an important role in promoting organizational innovation, which also supports an indirect positive role of Zhongyong thinking (Hu, 2017; Pian, 2019). However, further studies are warranted to verify the exact mechanisms underlying this process.

Our results showed that which tool was used to measure Zhongyong thinking did not have a significant moderating effect on the relationship between Zhongyong thinking and creativity. This might be because the scales used in the studies included in our meta-analysis were based on the Zhongyong Practical Thinking Scale developed by C. F. Yang and Chiu (1997), with each version having been reduced or refined (Q. Wei & Han, 2022). Therefore, the measured variable has the same connotation and the reliability of different measurement tools improves the comparability of the research results. The nonsignificant moderating effect of publication status showed that the research on the relationship between Zhongyong thinking and creativity was consistent in both publication types.

Implications

Our meta-analysis results demonstrate that Zhongyong thinking was positively associated with creativity, which sheds light on current arguments about whether Zhongyong thinking can promote or inhibit creativity in modern Chinese society (see, e.g., Y. Wu, 2009; X. Yao et al., 2010; Y. Yao & Fan, 2014; H. Zhou et al., 2017). Our findings reveal the core connotation of Zhongyong thinking, change the traditional cognitive bias of a one-sided view of Zhongyong, and advance understanding of the role of rational Zhongyong thinking.

Moreover, the relationship of Zhongyong thinking and creativity is of particular relevance for work organizations. First, it can be inferred that stimulating employees’ comprehensive and integrated thinking abilities is conducive to improving their ability to adjust themselves and integrate different views according to the environment, which can contribute to organizational innovation and ultimately improve organizational performance. Therefore, managers of enterprises need to strengthen guidance related to Zhongyong values by incorporating employees’ thinking modes and values into organizational construction. Second, the present meta-analysis of Zhongyong thinking and innovation in the Chinese cultural context supports the role of Zhongyong culture in promoting innovation, which can broaden cross-cultural research on the impact mechanism of organizational culture on organizational innovation and can enrich the diversity of organizational culture. Third, previous studies found that in organizations, Zhongyong culture, which is derived from traditional Chinese culture, is at odds with the spirit of excellence in the pursuit of innovation in modern organizations, so there is a divergence and contradiction between Chinese traditional culture and modern organizational culture (Y. Zhang, 2009). The results of our meta-analysis showed support for the positive role of Zhongyong culture in innovation, which provides an important reference for how to deal with the relationship between tradition and modernity within organizations during this period of great organizational change in modern society worldwide.

Limitations and Future Research Directions

There are some limitations in this study: First, Zhongyong thinking and creativity are two complex concepts. In this study we discussed the relationship between Zhongyong thinking and general creativity only, but creativity can be further subdivided. For example, the relationships between Zhongyong thinking and both individual and group creativity could be discussed for further clarity. Second, creativity and innovation behavior are related but different concepts (Yao et al., 2010). Creativity refers to the generation of new ideas, whereas innovation behavior is the activity of applying new ideas to new products and new production processes (Sarooghi et al., 2015). The transition from creativity to innovation is complex and may be influenced by Zhongyong thinking (X. Yao et al., 2010). Therefore, future studies should distinguish between creativity and innovation and could explore the relationship between Zhongyong thinking and these two concepts. Third, although we assumed that different measurement tools of Zhongyong thinking had the same connotation, in future research the correspondence or convergent validity of these diverse measures still need to be established from empirical evidence with the same sample. Fourth, the moderator analyses that we applied are not comprehensive. Some factors that may affect the relationship between Zhongyong thinking and creativity have not been measured. Therefore, more possible moderators need to be explored in future research. For example, one study found that Zhongyong thinking can positively predict incremental innovation and negatively predict radical innovation (Du et al., 2018). Therefore, the type of innovation may moderate the relationship. Finally, the results of a meta-analysis cannot reveal the causal relationship between Zhongyong thinking and creativity. Future studies could further explore this causal relationship by conducting longitudinal research.

Conclusion

In conclusion, we found a significant positive correlation between Zhongyong thinking and creativity in our meta-analysis. This result verified the relevant theoretical assumptions about Zhongyong thinking and creativity, which helps to clarify previous misunderstandings about Zhongyong thinking and also contributes to understanding of how to improve creativity from the perspective of Zhongyong thinking. In addition, the relationship between Zhongyong thinking and creativity was influenced by the creativity measurement method used, but was not moderated by the measurement tool of Zhongyong thinking or the publication status of the research.

References

This research was funded by the National Social Science Fund of China (14CSH038).

Qingwang Wei, Department of Psychology, Renmin University of China, No. 59, Zhongguancun Street, Haidian District, Beijing, People’s Republic of China. Email: [email protected]