Next Article in Journal
Study on the Characteristics of Coal Ultrasonic Response during Loading and Its Influence Mechanism
Next Article in Special Issue
Determinants of College Students’ Actual Use of AI-Based Systems: An Extension of the Technology Acceptance Model
Previous Article in Journal
Resilience of Public Open Spaces to Earthquakes: A Case Study of Chongqing, China
Previous Article in Special Issue
Sustainable Employability: Precariousness, Capabilities, and Functioning of Special Education Teachers in Namibia
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 15
Issue 2
10.3390/su15021094
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Does Help-Seeking Message Content Impact Online Charitable Behavior? A Qualitative Comparative Analysis Based on 40 Waterdrop Projects

by
Yanzhi Liu
,
Rong Cao
* and
Zheng Wang
College of Public Management, Northwest University, Xi’an 710127, China
*
Author to whom correspondence should be addressed.
Sustainability 2023, 15(2), 1094; https://doi.org/10.3390/su15021094
Submission received: 19 November 2022 / Revised: 3 January 2023 / Accepted: 4 January 2023 / Published: 6 January 2023
(This article belongs to the Special Issue Advances in Sustainable Psychology and Behavior: Managing and Developing People within the External and Organizational Environment)
Browse Figures
Versions Notes

Abstract

:
A help-seeking message is composed of abundant types of content; therefore, it is unsuitable for analysis by the traditional methods that assume that variables are independent of one another. To address this problem, we introduced qualitative comparative analysis (QCA) to explore the synergistic effects of help-seeking message content on online charitable behavior. Crisp-set QCA and fuzzy-set QCA were both used to analyze qualitative and quantitative data from 40 Waterdrop projects. To analyze the qualitative data, three members of our research team intensively and separately read a large number of help-seeking messages, analyzed and summarized the main content referring to previous studies on charitable donation, extracted rational appeals, positive emotions, negative emotions, moral appeals, and the economic condition as condition variables, and finally determined the coding rules collaboratively. The necessity analysis results show that moral appeals and rational appeals are necessary conditions for online charitable behavior. The sufficiency analysis results show that there are three configurations impacting online charitable behavior. This study can help inspire future studies shifting from a traditional perspective to a configuration perspective and help seekers obtain more charitable donations.

1. Introduction

Charitable behavior refers to the altruistic act of providing sponsorship to a community or individuals in need without any monetary rewards. It is needed to address a wide range of social issues, including poverty, education, medicine, environment, and culture [1], and it contributes to promoting social justice, strengthening social morality, and thus promoting sustainable development of the society. At present, with the rapid development, popularization, and application of Internet technology, an increasing number of Chinese people are engaging in online charitable behavior through Internet platforms such as Waterdrop, Easily-Chuao, Sina Micro Charity, and Tencent Charity. Since Internet platforms have a fast transmission speed, are easy to use, and have a high fundraising efficiency, they have shown strong potential in promoting individuals’ online charitable behavior and have received increasing attention as an emerging and considerable field. However, plenty of online charity projects have failed to achieve their target fundraising amount within a stipulated period [2]. Based on a survey of 13,119 projects from the Sina Micro Charity in China, the average fundraising success rate is merely 16.98% [3]. Therefore, it is important to identify the factors influencing individuals’ online charitable behavior and provide practical implications regarding how such behavior can be encouraged.
Compared with traditional charitable behavior, online charitable behavior is unique in that it is mediated by Internet platforms. In this situation, the donor and the help-seeker may be strangers to each other, and the help-seeking message can act as a direct communication bridge between them. Through the complex ideographic system composed of abundant help-seeking message content, help-seekers can guide donors to achieve value recognition, create emotional resonance, and make online charitable donations by stating objective facts (e.g., poverty due to illness), expressing subjective experience (e.g., pain), and expressing instinctive desires (e.g., to survive and be well). Therefore, it is critical to investigate which types of help-seeking messages are more conducive to promoting online charitable behavior.

2. Theoretical Framework

2.1. Literature Review

2.1.1. Online Charitable Behavior: Conceptualization and Definition

Although charitable behavior is of long-standing interest in social science, the study of online charitable behavior as an emerging field is still in an exploratory stage. Currently, there is varied use of the concept of charitable behavior relevant to Internet platforms in academic research. The terms “online giving” [4,5], “online charitable behavior” [6], “online donor behavior” [7], “online charitable giving” [3], “online donations” [8,9,10], and “online charitable donations” [11] are in common use. This study adopts the concept of online charitable behavior and does not strictly distinguish it from other concepts. Whatever concept is used, it does not prevent any understanding of this particular form of charitable behavior. Drawing on Bekkers’ classic definition of charitable behavior [12], this study defines online charitable behavior as the behavior of individuals who voluntarily provide property or services to others or nonprofit organizations through the Internet platform.

2.1.2. Factors Influencing Online Charitable Behavior

Previous studies have presented a trend of interdisciplinary research, including psychology, sociology, and public management, which primarily covers micro, medium, and macro levels, to address the practical problem of how individuals’ online charitable behavior can be increased.
At the micro level, the psychology of donors, the characteristics of help-seekers, and the social relationship between the donor and the help-seeker are significantly associated with online charitable behavior. For example, donors prefer to make online charitable donations to help-seekers with the same surname due to kin selection [9], sexual selection drives donors to donate more when facing the opposite sex on the Internet [5], and donors tend to make online donations to those who at least share an acquaintance with the help-seeker [13]. At the medium level, the qualities of Internet or platform service are important factors influencing online charitable behavior. For example, the number of new donors attracted by nonprofit Internet sites is directly related to their accessibility, accountability, interaction, and empowerment [14]. Internet technical features such as security and privacy influence individuals’ general attitudes and behaviors toward online charitable donation [15]. Humanitarian projects and Internet technology significantly affect donors’ attitudes towards online donation [16]. The visibility, reputation, and transparency of platforms also influence online charitable behavior [7]. Setting default donation amounts can effectively enhance Internet charitable donation levels, but such default settings must take into account individuals’ previous preferences for charitable behavior choices [17]. At the macro level, researchers examined the effectiveness and inadequacy of charity laws and policies in regulating or motivating online charitable behavior. For example, Ke posited that some Internet platforms still escape from the legal regulation and government supervision in China [18]. Shi, Fan and Gao, based on the Chinese context, discussed the comprehensiveness of religious policy content, the feasibility of implementation, and antecedents that influence the impact of policy implementation [19].

2.1.3. Studies on the Relationship between Help-Seeking Message Content and Online Charitable Behavior

Some scholars have realized the role of help-seeking messages in bridging charitable resources and charitable needs and have used regression analysis to examine the impact of help-seeking message content on online charitable behavior. For example, Chen and Yue summarized the unstructured textual language of the Sina Micro Charity and built a regression model from three dimensions based on grounded theory: the story development process, specific details, and requests for help [20]. Majumdar and Bose, referring to Aristotle’s persuasion theory, also used regression analysis to explore the influence of rational appeals, emotional appeals, credibility appeals, and other help-seeking message content on online charitable behavior [21]. Fuguo and Bingyu, collecting 4903 projects from the Tencent Charity, studied the relationships between the presence of the patient’s occupation, monetary evidence, negative emotions, positive emotions, and online charitable behavior using multiple regression analysis [22]. These studies provide a solid foundation for designing help-seeking messages to increase online charitable behavior, but there are still issues to be addressed.
To begin with, traditional methods (such as linear regression, structural equation model, and factor analysis) use economic marginal analysis as a reference and strive for optimal distinctiveness [23]. However, as the saying goes, “All roads lead to Rome”. The same destination can often be reached by different routes and there can be multiple reasons why a particular outcome may or may not occur [24]. In other words, outcome Y may be caused by A, by both A and B, or by a combination of B, C, and D. This means that the reasons for donors’ online charitable behavior are complex and varied. Secondly, traditional analysis methods assume that the reasons for the occurrence or non-occurrence of a specific outcome are uniform. For example, if positive emotions have a positive impact on charitable behavior [25], the more positive the emotional state is, the higher the level of charitable donation will be. Conversely, the more negative the emotional state is, the lower the level of charitable donation will be. However, donors often exhibit high levels of charitable donation when their emotion is negative [22,26,27], demonstrating that the same independent variable has contradictory relationships with the dependent variable in different studies. Although the establishment of interaction effects can improve the explanatory power of a theoretical model for empirical observation to a certain extent, it can only handle three interactions at most [28]. Finally, traditional analysis methods assume that multiple independent variables are independent of one another and focus on the net effects of a separate independent variable on the dependent variable, ignoring the interdependence relationships between variables [29]. In reality, a help-seeking message is an organic whole made up of various components that have complex complementary, substitutional, and suppressional relationships. Therefore, it is not suitable to conduct this research through traditional methods.
Based on the above analysis, this study aims to introduce a qualitative comparative analysis (QCA) approach with the basic assumptions of equifinality, causal asymmetry, and multiple conjunctural causations [24] and explore the impact of help-seeking message content on online charitable behavior from a configurational perspective.

2.2. Research Model

The first step in a QCA study is usually to select appropriate condition and outcome variables that depend on the research question and develop a reasonable configuration model. The selection of condition variables should be based on the iterative dialogue between case materials and theoretical research [30]. On the one hand, the setting of condition variables should conform to the internal structure of case materials. That is to say, the material for every case should contain condition variables available for coding. On the other hand, the selection of condition variables must also be built on a solid theoretical basis. In other words, corresponding theoretical literature or previous studies have demonstrated the rationality of the existence of the condition variable and have paid attention to its impact on the outcome of the case. Accordingly, this study comprehensively adopted the phenomenon summary method and the literature induction method [31]. During the condition selection process, our research team first deeply interpreted and summarized the core content of the help-seeking messages collected from Waterdrop, then referred to the existing literature on charitable donation, and finally extracted five condition variables.
(1)
Rational appeals. Rational appeals emphasize the use of objective facts and rational logic analysis to prove the authenticity of the help-seeking message to donors. Because of the information asymmetry problem in the Internet space, help seekers may conceal their real family’s economic condition and exaggerate medical expenses, so donors are often skeptical about the authenticity of the messages they receive. In this situation, objective, detailed, and reasonable statements can help to increase donors’ perceptions of credibility and build trust in the help-seeking message, which in turn promotes their charitable intentions and behavior. Previous research has also shown that in-depth narratives, clear statements of need, and longer stories help to motivate charitable behavior [21,22].
(2)
Emotional appeals. Emotional appeals (including positive emotions and negative emotions) are fundraising techniques that use emotional materials such as words, images, or videos to evoke positive or negative emotions in donors, thereby inspiring them to make charitable donations. Based on the theory of emotional contagion, individuals tend to imitate the expressions, voices, postures, and movements of their peers during social interactions unconsciously to stay in sync with their emotions [32]. Thus, help seekers’ emotional expression can trigger similar emotions in donors. That is to say, help seekers can infect the emotions of donors with emotional materials in a help-seeking message. Previous studies have also shown that negative emotional materials can lead to loss aversion in individuals and positive emotions are more likely to trigger charitable behavior than negative emotions [25,33]. However, according to the negative state release model, altruistic behaviors have the function of self-satisfaction. Individuals in negative emotional states will carry out altruistic behavior to repair their emotions, while positive emotions reduce the need for individuals to relieve negative states through altruistic behavior. Therefore, negative emotions can stimulate altruistic behavior more strongly than positive emotions [26,34].
(3)
Economic condition. The economic condition expresses the need for money by highlighting household income, medical expenses, or other expenditures. Previous research has found that exposure to money-related images or other materials can significantly reduce individuals’ willingness to contribute to a charity and actual charitable behavior [35,36,37]. As a kind of social resource, money can well meet the demand of individuals, triggering feelings of self-sufficiency and reducing individuals’ concern for social goals and others [38,39]. Exposure to monetary cues may also lead individuals to think of social interactions in a more utilitarian way. Those exposed to monetary cues are inclined to believe that correcting the natural elimination caused by survival competition through charity is not conducive to social progress, thus discouraging charitable behavior [40].
(4)
Moral appeals. By highlighting moral values, moral appeals can stimulate a sense of obligation and exhort individuals to behave morally. Based on norm activation theory, altruistic behavior occurs when individuals are aware of their moral obligations [41]. Charitable behavior, as a socially recognized noble moral behavior or altruistic behavior, can be stimulated by the moral discourses quoted in help-seeking messages. Relevant studies have proved that moral appeals that emphasize intrinsic moral awareness and group norms are conducive to charitable behavior [42]. In the field of organ donation, moral appeals such as altruism can also increase an individual’s sense of obligation and commitment to donate, thus promoting organ donation [43].
On the basis of the literature review on factors affecting charitable behavior, this study developed a research model (see Figure 1 below), with online charitable behavior as an outcome variable, and rational appeals, positive emotions, negative emotions, moral appeals, and economic condition as condition variables. The research model helps to further identify what help-seeking message content is necessary for online charitable behavior and what configurations of help-seeking message content are sufficient for online charitable behavior.

3. Methods

QCA is a method based on Boolean algebra and set theory, which was proposed by Charles C. Ragin at the end of the 19th century. It is presupposed that social causality is nonlinear and causal factors are both interdependent and substitutable. Therefore, analysts should treat each case as a configuration of conditions and find multiple configurations of conditions that lead to the presence or absence of an outcome through case comparison [30].
Unlike traditional methods based on correlation, QCA is built on set theory, which analyzes the relationship between the condition set and the outcome set, and thus aids in the provision of critical and explicit information on social phenomena [44]. Set analysis mainly includes an analysis of the necessity of a single condition and the sufficiency of the configuration of multiple conditions. Regarding a necessary condition, the condition forms the superset of the outcome, so the outcome cannot present without the condition; regarding a sufficient condition, the condition forms a subset of the outcome, so the condition can lead to the outcome adequately [44]. However, such information is neglected by traditional methods. Researchers may find a set relation between two variables, but there may be no significant correlation between them [45].
Overall, QCA is a mixed method that combines the advantages of both qualitative and quantitative strategies [30]. On the one hand, it has a standardized procedure, using as much quantitative analysis as possible in the coding and computational aspects, which ensures the transparency and reproducibility of the research. On the other hand, it uses a cluster of cases with similar structured characteristics as the study object and is thus less subjective than qualitative analysis. Given the advantages of combining case-based methods and variable-oriented methods, QCA has been widely used in the social sciences, e.g., in public administration [46,47], education [48], and sociology [49].
QCA can be divided into crisp-set QCA (cs-QCA), multi-value QCA (mv-QCA), and fuzzy-set QCA (fs-QCA) according to data types. In particular, crisp-set QCA is used to analyze “non-membership” and “full membership” dichotomous data. If a condition presents itself, it is valued at 1; otherwise, it is valued at 0. Nonetheless, many variables in the complex reality cannot be clearly identified as present or absent. This problem can be solved by multi-value QCA or fuzzy-set QCA. Multi-value QCA is used to analyze multi-valued nominal data, while fuzzy-set QCA is used to process continuous data, allowing variables to obtain partial membership scores ranging from 0 to 1 [24].
Based on the standardized procedure of QCA [45], we developed the following procedure (shown in Figure 2): (1) based on the practical problem of how to increase online charitable behavior through the design of a help-seeking message, construct a theoretical framework referring to previous research; (2) collect 480 project data to form a case database and meticulously select 40 cases for further research according to the principles of similarity, comparability, and diversity; (3) use both crisp-set QCA and fuzzy-set QCA to measure and calibrate the condition and outcome variables; (4) conduct the necessity analysis, adequacy analysis, and robustness testing successively; and (5) construct a theoretical model on how help-seeking message content affects online charitable behavior to deepen and refine the findings of this study.

3.1. Data Collection and Case Selection

Waterdrop is one of the second batches of online fundraising platforms for charitable organizations or individuals in need designated by the Ministry of Civil Affairs of China. In recent years, Waterdrop has enjoyed high popularity and recognition. According to a previous study, of the 513 Nanchong residents surveyed, 89.77% reported knowing about Waterdrop projects on QQ or WeChat, and after reading the help-seeking messages, 69.59% reported making online charitable donations, and 56.3% reported forwarding the help-seeking messages [50]. Thus, Waterdrop is a representative object in a study of online charitable behavior. The Waterdrop project is typically presented in the form of a link, which can be clicked on to view details about the help-seeker’s name, age, photos, family assets, target fundraising amount, status of fundraising, and a message telling the story of his or her family and other relevant details.
In contrast to the requirements of traditional methods on the number of cases, QCA pays more attention to the selection of cases. It follows theoretical sampling rather than random sampling in case selection. It requires researchers to carefully select cases based on the principles of similarity, comparability, and diversity [31]. In this study, the unified use of the Waterdrop project data can ensure maximum similarity and comparability between different cases. In terms of diversity, we first ranked the 480 cases in the database according to their fundraising rates and determined the highest (0.54) and lowest (0.10) values. We then divided the range into intervals of 0.10–0.20, 0.20–0.30, 0.30–0.40, 0.40–0.50, and 0.50–0.60, so that the primary cases would be evenly distributed in each interval. Finally, the cases were meticulously selected to cover the different condition variables (In QCA, case selection and condition selection are processes that are carried out repeatedly based on theoretical research and empirical knowledge. The different condition variables referred to here are rational appeals, positive emotions, negative emotions, economic condition, and moral appeals as cited in Section 2.2) and fundraising rates. QCA is best suited for studies with small and medium sample sizes ranging from 10 to 60 [51], so we ultimately chose 40 cases.

3.2. Measurement and Calibration

In conjunction with the real scenario of the cases, we applied both the crisp-set QCA and the fuzzy-set QCA. Since fuzzy-set QCA can concurrently analyze problems of kind and degree, we used it rather than crisp-set QCA as much as possible when measuring and calibrating the condition and outcome variables.

3.2.1. Outcome Variable

The outcome variable of this study is online charitable behavior. As previous research has demonstrated a significant “pull-up” effect of target fundraising amount on online charitable behavior [12,52], to ensure comparability across cases, we used the fundraising rate (the ratio of the online charitable donation amount to the target fundraising amount) to measure online charitable behavior. Fundraising success (calibrated code infinitely close to 1) represents a high level of charitable behavior, while failure (calibrated code infinitely close to 0) represents a low level of charitable behavior.

3.2.2. Condition Variable

QCA is a mixed method that combines the advantages of both qualitative and quantitative strategies [30]. As mentioned before, crisp-set QCA allows variables to assign values of 0 or 1, while fuzzy-set QCA allows variables to obtain partial membership scores ranging from 0 to 1. In practice, the researcher could make a judgment for each variable based on theoretical studies and experiential knowledge [24]. The coding of qualitative data is somewhat flexible for researchers to adapt to the research area, question, and context. As a result, this may inevitably result in a degree of subjectivity. To reduce the subjective bias and ensure the scientific nature of our study, we relied on teamwork. Three members of our team first determined the coding rules of the condition variables separately and then conducted a joint discussion to reach a consensus. In addition, we also referenced other similar studies involved in the coding and assignment rules for qualitative data [53,54]. The final rules in this study are as follows (Appendix A provides examples of this process).
Positive emotions are measured by the ratio of positive words to the length of the text, whereas negative emotions are measured by the ratio of negative words to the length of the text. Rational appeals are scored using the patient’s name, age, home address, disease name, and treatment process and are divided into 6-valued fuzzy sets, with a value of 0 representing that all conditions are not met, a value of 0.2 representing that one condition is met, a value of 0.4 representing that two conditions are met, a value of 0.6 representing that three conditions are met, a value of 0.8 representing that four conditions are met, and a value of 1 representing that five conditions are met. Moral appeals are measured by the presence or absence of moral discourse, which assigned a value of 1 if it occurs and a value of 0 if it does not. Economic condition is measured by the patient’s occupation, family members, household income, and treatment costs and is divided into 5-valued fuzzy sets, with a value of 0 representing that all conditions are not met, a value of 0.25 representing that one condition is met, a value of 0.5 representing that two conditions are met, a value of 0.75 representing that three conditions are met, and a value of 1 representing that all conditions are met. The coding rules of all variables are shown in Table 1.
To ensure that the raw data can be interpreted and has set significance, calibration is necessary before performing the analysis, which refers to the process of assigning a set membership score to a case. During calibration, three thresholds were set: full non-membership, the crossover point, and full membership [45]. This calibration method can minimize subjective bias. Based on the data types of the condition and outcome variables, we used fs-QCA3.0 to calibrate the quantitative data—positive emotions, negative emotions, and online charitable behavior—so as to convert them into fuzzy set membership scores. The calibration results show that the full non-membership, crossover point, and full membership values of positive emotions are 0.0000, 0.0070, and 0.0128, those of negative emotions are 0.0070, 0.0198, and 0.0354, and those of online charitable behavior are 0.2125, 0.3250, and 0.4200, respectively. The calibrated results for all variables are shown in Table 2.

4. Results

4.1. Necessity Analysis

We first tested whether single conditions constitute necessary conditions for online charitable behavior. When a condition is always present when an outcome occurs, it is a necessary condition for the outcome. The necessary condition is tested by a consistency level greater than 0.9 [45].
The results (see Table 3) show that the level of consistency of moral appeals is greater than 0.9, demonstrating moral appeals are necessary for online charitable behavior. This indicates that help-seeking messages that are effective in motivating online charitable behavior inevitably contain moral discourse. If a message does not include any moral discourse, it is usually difficult to achieve the desired results. This finding is in line with the theoretical views of previous scholars [43], which suggests that the quote of moral discourse in the message can effectively integrate the needs of donors with social morality, thereby stimulating a sense of moral obligation in donors and triggering charitable behavior. The reason why moral appeals affect charitable behavior is that individuals have the psychological tendency to perceive, abide by, and spread moral norms. This tendency is not only the product of adaptive selection in biology, but also the result of the conscious creation of human society. On the one hand, the biological traits beneficial to survival and reproduction are preserved in the long-term natural selection process, forming the physiological basis of individual moral emotions and behaviors. On the other hand, human society has designed and formulated moral norms to maintain cooperation and common interests, which constitutes the social basis of individual moral emotions and behaviors [55]. Charitable behavior itself and social morality are complementary and mutually reinforcing. The widespread participation of the public in charitable behavior helps to raise the level of social morality, while the transmission and promotion of social morality contribute to charitable behavior in a “silent” manner, ultimately forming an upward social atmosphere toward good.
The level of consistency of rational appeals is as high as 0.8961, which is approximately equal to 0.9, indicating that rational appeals can also be considered as a necessary condition for online charitable behavior, which is consistent with previous studies [21,22]. Donors are cautious about the truth of information they receive because of the asymmetry of information in cyberspace. Those help-seeking messages that are objective, thorough, and logical can help donors to obtain trust, thus improving their desire and behavior to donate online.
The level of consistency of positive emotions, negative emotions, and economic condition, however, do not meet the minimum triggering requirement for donor giving, suggesting that they must be combined with other help-seeking message content to effectively motivate online charitable behavior. Therefore, it is necessary to further analyze the relationship between help-seeking message content and online charitable behavior using configuration analysis.

4.2. Sufficiency Analysis

Sufficiency analysis attempts to explore the adequacy of the outcome caused by different configurations of multiple conditions. In this way, it needs to set the column number and consistency index and screen out the conditions that can fully explain the results. The frequency threshold is determined by the sample size, which is generally 1 for small to medium-sized samples, and the consistency threshold is generally above 0.75 [24]. In this study, we set the frequency threshold at 1 based on the sample size, the raw consistency threshold at 0.8228 based on standardized configuration analysis, and the PRI consistency threshold at 0.75 as per usual.
The results of QCA include three solutions: a complex solution, an intermediate solution, and a parsimonious solution. In practical research, the intermediate solution with moderate complexity is usually the first choice for reporting and interpretation [31]. Therefore, we focused on the analysis based on the intermediate solution. To further explore the core conditions and peripheral conditions of each configuration, we identified the condition variables appearing in the parsimonious solution as core conditions, and those appearing only in the intermediate solution and excluded from the parsimonious solution as peripheral conditions [56]. The results (see Table 4) show that there are three configurations for online charitable behavior. The overall solution consistency of the single solution and the overall solution of the three configurations are both higher than the minimum acceptable standard of 0.75, which proves that the three configurations are sufficient to effectively stimulate online charitable behavior. The overall solution coverage is 0.6349, indicating that the three configurations can explain 63.49% of the 40 cases.
In Configuration 1, the presence of moral appeals, the absence of negative emotions, and the absence of economic condition play core roles, while the presence of rational strategies plays a peripheral role. The consistency level of this configuration pathway is 0.8297, the raw coverage is 0.3293, and the unique coverage is 0.1326, indicating that this pathway can explain about 32.93% of the 40 cases. About 13.26% of the 40 cases can only be explained by this pathway. This result suggests that an objective and detailed account of the background of help-seeking based on the inclusion of moral discourse can synergistically motivate online charitable behavior by stimulating potential donors’ sense of moral obligation [43] and enhancing the perception of message authenticity [21,22]. At the same time, reducing the frequency of negative words can help increase online charitable behavior, because help-seeking messages related to serious diseases usually contain the symptoms of the disease and the devastating impact on the family, which can easily activate individuals’ fear and non-adaptive defense mechanisms [57]. Thus, information avoidance behavior is generated, which further leads to a decrease in the opportunity to receive online charitable donations. Meanwhile, the reduction of the patient’s occupation, income, family members, treatment costs, and other money-related content can prevent individualistic tendencies and increase the likelihood of online charitable behavior [35,36].
In Configuration 2, the presence of positive emotions and the presence of moral appeals play core roles, while the presence of rational appeals and the absence of economic condition play peripheral roles. The consistency level of this configuration pathway is 0.8607, the raw coverage is 0.4579, and the unique coverage is 0.2612, indicating that this pathway can explain about 45.79% of the 40 cases. About 26.12% of the 40 cases can only be explained by this pathway. This result indicates that using positive words to express emotions, including moral discourse, and employing objective and detailed statements simultaneously when seeking help are effective in stimulating online charitable behavior. This is because messages containing positive words can awaken donors’ positive emotions, broaden their attention spans, and increase their perceived value of the rewards of altruistic behavior [58]. At the same time, money is more than just a neutral medium of exchange for goods; it also represents rich psychological experiences of success, power, social approval, and security, as well as moral guilt and social alienation, all of which can alter individuals’ cognitive, emotional, and behavioral responses. As stated earlier, reducing money-related cues can lessen potential donors’ sense of self-sufficiency [38,39] and level of utilitarian values [40,59], thereby increasing their willingness and behavior to donate online.
In Configuration 3, the presence of positive emotions and the presence of moral appeals play core roles, while the presence of rational appeals and the absence of negative emotions play peripheral roles. The consistency level of this configuration pathway is 0.8505, the raw coverage is 0.2410, and the unique coverage is 0.0444, indicating that this pathway can explain about 24.10% of the 40 cases. About 4.44% of the 40 cases can only be explained by this pathway. This result suggests that an objective and detailed account, based on the use of positive words to express emotions and the inclusion of moral discourse, is conducive to promoting online charitable behavior by simultaneously extending the donors’ positive attention span [58], invoking a sense of moral obligation [43], and generating trust in help-seeking messages [21,22].
By comparing the three configurations, we can see that, in all configurations, moral appeals are a core condition and rational appeals are a peripheral condition, which further proves that they both, and especially moral appeals, are indispensible in stimulating online charitable behavior. Varying help-seeking message content can act as a complement through these conditions’ presence or absence and can play a core or peripheral role, forming synergistic effects on online charitable behavior. Specifically, “the core role of the presence of positive emotions and the peripheral role of the absence of economic condition,” “the core role of the presence of positive emotions and the peripheral role of the absence of negative emotions,” and “the core role of the absence of negative emotions and economic condition” are mutually substitutive. “The peripheral role of the absence of economic condition” and “the peripheral role of the absence of negative emotions” are substitutes for each other. Negative emotions and economic condition have inhibitory effects, and their absence combined with other help-seeking message content can effectively drive the online charitable behavior of donors.
In addition, the three configurations in this study can explain 63.49% of the 40 cases. The reason why the rest of the cases are not covered may be that there are other condition variables simultaneously affecting online charitable behavior. For example, a Waterdrop project often includes not only content factors but also images [21,60], proofs [61], fundraising dynamics [12,52], project reviews [62], and other non-content factors that are closely related to online charitable behavior. They can also constitute a larger organic whole that is equally capable of forming other pathways. However, since the number of condition variables should be between 4 and 7 in research with small and medium-sized samples [24], these non-content factors were not considered in the present study.

4.3. Robustness Testing

Robustness testing can be achieved by making reasonable adjustments to relevant parameter settings, e.g., by changing calibration, altering consistency levels, and dropping or adding cases. If the parameter adjustment does not lead to substantial changes in the number of configuration pathways, condition distribution arrangements, consistency levels, and coverage, the research results are robust [45]. In this regard, we adjusted the original consistency threshold to 0.85 and conducted the analysis again. The results show that none of the three pathways changed, and the overall consistency level and coverage rate did not change significantly, indicating the results of this study are robust.

5. Discussion and Conclusions

5.1. Conclusions

This study explored the impact of help-seeking message content on online charitable behavior based on a QCA method using 40 Waterdrop projects. The results showed the following:
(1)
Moral appeals are a necessary condition for online charitable behavior.
(2)
Rational appeals are a necessary condition for online charitable behavior.
(3)
There are three equivalent pathways for the influence of help-seeking message content on online charitable behavior. In Configuration 1, the presence of moral appeals, the absence of negative emotions, and the absence of economic condition play core roles, while the presence of rational appeals plays a peripheral role. In Configuration 2, the presence of positive emotions and the presence of moral appeals play core roles, while the presence of rational appeals and the absence of economic condition play peripheral roles. In Configuration 3, the presence of positive emotions and the presence of moral appeals play core roles, while the presence of rational appeals and the absence of negative emotions play peripheral roles.
Upon the above conclusions, we further referred to the Stimulus-Organism-Response (SOR) theory [63] to develop a process model, which reveals the mechanism of how help-seeking message content influences online charitable behavior (shown in Figure 3).
Firstly, by compiling rational appeals, emotional appeals, economic condition, moral appeals, and other help-seeking message content, help-seekers selectively construct stories with a multitude of details, thus generating a variety of external environmental stimuli. Secondly, when donors are exposed to these stimuli in social media such as WeChat, Weibo, and QQ, they will experience internal changes, both cognitively in terms of risk perception and utilitarian values and emotionally in terms of trust, fear, sympathy, morality, and self-sufficiency [64]. Finally, these changes further drive them to decide whether to donate or not or how much they should donate. This process unfolds step by step, forming a complete logic chain from message stimuli to internal state change and then to behavioral response, exhibiting highly complex features of equifinality, causal asymmetry, and multiple conjunctural causations.

5.2. Practical Implications

The necessity analysis results implicate that help seekers can evoke the goodness of human nature by including moral discourse when writing the message so that donors can consciously externalize the online charitable behavior. They can also explain the background of help-seeking as rationally and minutely as possible so that donors’ trust in the message will be enhanced.
The sufficiency analysis indicates that help seekers need to take a holistic view and avoid being restricted to a single type of content, such as rational appeals, emotional appeals, moral appeals, and economic condition. They can remain open and flexible according to the diversified configurations of help-seeking message content. In terms of the specific details, help seekers need to constantly innovate the narrative techniques based on the inclusion of moral discourse, the disclosure of personal information such as the patient’s name, age, and home address, the inclusion of objective details regarding the disease and treatment process, expressions of positive emotions rather than negative ones, and the reduction of monetary cues such as the patient’s occupation, family members, household income, and treatment costs.

5.3. Contributions and Limitations

The theoretical contributions of this study are as follows: First, it investigated the causal complexity of help-seeking message content in motivating online charitable behavior using QCA, which inspires future studies to shift from the traditional perspective based on the sole optimum distinctiveness, uniform causality, and net effects assumption to the configuration perspective of equality, causal asymmetry, and multiple conjunctural causations. Second, a process model based on the SOR theory was developed, which drew a logical chain from the donor receiving message stimuli to their internal changes and then to their behavioral response, which is useful in clarifying the mechanism by which help-seeking message content influences online charitable behavior. Third, the acquisition and use of the behavioral data from Waterdrop projects help to better understand online charitable behavior in the real world. This study also has practical value. It focused on the characteristics of help-seeking messages and explored configurations that achieve higher fundraising performance, which is conducive to designing effective strategies for help seekers.
This study, however, has some limitations. First, the 40 Waterdrop cases may limit the external validity of the findings. To enhance the credibility and universality of the conclusions, future studies need to expand the sample size and combine it with data from other Internet platforms for verification. Secondly, the fuzzy set calibration process of qualitative data lacks objective and unified standards, so the coding of the condition variables in this study is subjective to a certain extent. Future research should be combined with other methods to revise and improve these conclusions. Lastly, the impact of help-seeking message content on online charitable behavior was only considered in a static context, and future research could introduce Temporal Qualitative Comparative Analysis (TQCA) to investigate their dynamic relationship.

Author Contributions

All authors made a significant contribution to the work reported. Conceptualization, Y.L. and R.C.; methodology, Y.L. and R.C.; software, Y.L.; validation, R.C. and Y.L.; formal analysis, R.C.; investigation, Y.L.; resources, Y.L.; data curation, Y.L., R.C. and Z.W.; writing—original draft preparation, Y.L.; writing—review and editing, Z.W. and R.C.; visualization, Z.W.; supervision, Z.W.; project administration, Z.W.; funding acquisition, Z.W. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Social Science Foundation of Shaanxi Province of China, grant number 2018P05.

Institutional Review Board Statement

Our study involves data that have been made public by the dataset owner (Waterdrop).

Informed Consent Statement

According to the relevant laws, regulations, and privacy policy of Waterdrop, it is not necessary to obtain the authorization and consent of fundraising initiators when an academic study is conducted for the public interest and has anonymized the personal information of fundraising initiators. Therefore, our study is deemed as meeting the institutional requirements and is exempt from ethical requirements.

Data Availability Statement

The datasets analyzed during the current study are not publicly available due to personal privacy but are available from the corresponding author on reasonable request.

Acknowledgments

We would like to thank Waterdrop for sharing the data.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Appendix A. Examples of Help-Seeking Messages and the Coding Process of Condition Variables

During selection of condition variables and determining their assignment rules, three members of the research team intensively and separately read a large number of help-seeking messages, analyzed and summarized the main content referring to previous studies on charitable donation, then extracted rational appeals, positive emotions, negative emotions, moral appeals, and economic condition as condition variables, and finally determined the assignment rules collaboratively. The following three translated help-seeking messages and the marks will help readers fully understand the process of this study:
Case 1:[M1] [慈2]
Dear kind and loving people! Hello! I am sorry to appear in your friend circle in such a way. My wife is XXX, she is 39 years old, and our home address is XXX. Our family is not rich but has been healthy and very happy in the past. However, “fortune is as unpredictable as the weather”. My wife was unfortunately found to be suffering from a breast malignant tumor, a lung secondary malignant tumor, and a bone secondary malignant tumor on February 2 this year. The doctor told me that my wife is in a very serious condition; she needs to be given chemotherapy immediately to control the disease, and the medical expenditure is very large. The news is undoubtedly very bad for our family. In fact, our family is very poor. I and my wife have two daughters who are still in school and incur a lot of expenses. At the same time, my wife is disabled, and she is usually at home to take care of our children. Now, we don't know what to do in the face of such huge treatment costs. My wife is still young, and our children cannot live without their mother. Thus, we decided to ask for help through Waterdrop. We hope that you can help my wife get through this difficult time and continue to fight the disease. Every forwarding from you is a hope for our family!
The follow-up medical expenses are really a bit too high, and our family really cannot afford such a large treatment cost. There is no choice for us but to plead with the community of kind-hearted people to save my wife! I hope you can help her so that she can be treated smoothly, and recover as soon as possible. We will remember your kindness, and we will be sure to repay it! I hope that all the kind-hearted people in society will help our family get through this difficult time and create a miracle of life. Even if it is a penny or a blessing, it can bring us great hope and encouragement. Our family kowtow to thank! “Good people live in peace! The merit is immense”!
Case 2:
Dear friends, family, and kind-hearted people! I am sorry to disturb you. This is my father, whose name is XXX, and he is 75 years old. In July and August 2020, my father's weight started to drop. After he received the new vaccination in November, his resistance dropped, and he was suffering from various symptoms, including vague pain in his abdomen. He thought it was caused by an inadvertent fall and held on to a sliver of luck, so he never went to the hospital. Later, the pain was so severe and unbearable that I took my dad to the Zhejiang University School of Medicine for a checkup. After a few days of anxious waiting, the diagnosis report was settled, and the final result was unacceptable. Dad was unfortunately diagnosed with advanced gastric cancer!
We couldn't believe that Dad had such a serious disease and that he, who was still alive, would collapse so quickly. Faced with this crazy disease and heavy medical expenses, my dad wanted to give up his treatment. As a son, my heart cut like a knife as I watched my frail father struggling on the hospital bed. And I would desperately grit my teeth to persevere, even if it was difficult.
We are a poor family, and we never expected to encounter this disaster! In order to bear the cost of treatment, we are at the end of our rope and have no choice but to turn to Waterdrop to raise money. Please help us, kind-hearted people, so that my tormented father can receive the treatment. As long as there is a ray of hope, I will never give up! Please stick with us and be a strong fighter all the time. Whether it is your generosity or your help in forwarding the message, it is a great help to our family. Allow more people to see us, and thus there will be more hope. Every forward you make is vital to us, and every forward is a great help to us! We desperately need your help!
Case 3:
My name is XXX, and I want to ask for help for my mother. Our home address is XXX. My mother was in a car accident while working in front of the Tiantai Shuyuan construction site. She was hit by a motorcycle traveling from north to south. She was dragged about 10 meters. A few of her coworkers rushed to call the 120 emergency vehicles and sent her to the West Hospital of Zibo Central Hospital. After the accident that day at the Zibo Center, through various checks, the doctor said the injury was very serious. At any time, there is [life-threatening], and the surgery must be performed immediately. Hurry up and arrange surgery for us. Because of the injuries which were [very serious], my mother has underwent a [spleen removal] [small intestine cut off a section] [liver stitches] [lung removal piece] surgery. The doctor said my mother's condition is very serious. Now she must be taken cared intensively, and the daily cost is at least 20,000 yuan. My mother has to stay in the ICU for at least 20 days. At any time there is [life threatening]. In just three days, we have spent more than 100,000 yuan. My mother's workplace has now bailed us out for 20,000 yuan. The owner of the motorcycle that hit my mother gave us 13,000 yuan. The doctor said the follow-up expenses are still very large. If the condition is stable, we will expend more than 300,000 yuan. If the condition worsens, the cost will be more. My mom and dad only have part-time jobs to earn money. Their monthly salary is no more than 4000 yuan. Now, my dad is in the hospital to take care of my mom. Thus, our family has no income. And I am still in college. There is no income.
The hospital knows that our family is poor, and the doctors and nurses have taken extra care of us and tried to help us reduce our expenses in any way they could. I am grateful to them from the bottom of my heart. Even if it is only a forwarding! Even if it is only a word of cheer! It is a great help for our family. It's my hope. I desperately need your help. Every forwarding from you is vital to me, and every forwarding is a great help to me. Thank you all! We desperately need your help! I hope kind-hearted people reach out to help! One more forward is one more hope. “Relay to light up the light of hope for life”!
Note: Blue highlights represent objective, rational statements of fact, green represents positive emotional expressions, pink represents negative emotional expressions, yellow represents money-related information, and gray represents an appeal to morality. Underline indicates words denoting negative emotion, and italics indicates words denoting positive emotion.
The name and address have been anonymized to protect the privacy of help seekers.
The translated messages may differ from the original meaning, and thus the marks may seem somewhat biased. Nevertheless, they could clearly present our work on the whole.

References

  1. Park, E.Y.; Rhee, J.H. Who clicks on online donation? Understanding the characteristics of SNS users during participation in online campaigns. Sustainability 2019, 11, 3674. [Google Scholar] [CrossRef] [Green Version]
  2. Chen, Y.; Dai, R.; Yao, J.; Li, Y. Donate time or money? The determinants of donation intention in online crowdfunding. Sustainability 2019, 11, 4269. [Google Scholar] [CrossRef] [Green Version]
  3. Chen, X.; Song, D.; Shou, Y.; Pan, Y. Altruism or social motives? Evidence from online charitable giving in China. Enterp. Inf. Syst. 2022, 16, 566–588. [Google Scholar] [CrossRef]
  4. Dean, J. Student perceptions and experiences of charity on social media: The authenticity of offline networks in online giving. Volunt. Sect. Rev. 2020, 11, 41–57. [Google Scholar] [CrossRef]
  5. Raihani, N.J.; Smith, S. Competitive helping in online giving. Curr. Biol. 2015, 25, 1183–1186. [Google Scholar] [CrossRef] [Green Version]
  6. Farrelly, D.; Bennett, M. Empathy leads to increased online charitable behaviour when time is the currency. J. Community Appl. Soc. Psychol. 2017, 28, 42–46. [Google Scholar] [CrossRef]
  7. Shier, M.L.; Handy, F. Understanding online donor behavior: The role of donor characteristics, perceptions of the internet, website and program, and influence from social networks. Int. J. Nonprofit Volunt. Sect. Mark. 2012, 17, 219–230. [Google Scholar] [CrossRef]
  8. Wang, P.; Li, J.; Ma, Y.; Jiang, Z. Quantifying the endogeneity in online donations. Entropy 2021, 23, 1667. [Google Scholar] [CrossRef]
  9. Sisco, M.R.; Weber, E.U. Examining charitable giving in real-world online donations. Nat. Commun. 2019, 10, 35–57. [Google Scholar] [CrossRef] [Green Version]
  10. Saunders, T.J.; Taylor, A.H.; Atkinson, Q.D. No evidence that a range of artificial monitoring cues influence online donations to charity in an Mturk sample. R. Soc. Open Sci. 2016, 3, 150710. [Google Scholar] [CrossRef]
  11. Erceg, N.; Burghart, M.; Cottone, A.; Lorimer, J.; Manku, K.; Pütz, H.; Vlašiček, D.; Willems, M. The effect of moral congruence of calls to action and salient social norms on online charitable donations: A protocol study. Front. Psychol. 2018, 9, 1913. [Google Scholar] [CrossRef]
  12. Bekkers, R.; Wiepking, P. A literature review of empirical studies of philanthropy: Eight mechanisms that drive charitable giving. Nonprofit Volunt. Sect. Q. 2011, 40, 924–973. [Google Scholar] [CrossRef]
  13. Zhu, H.; Yin, K.; Yang, L.H. Effects of beneficiaries’ facial expressions and donor-beneficiary relationship on donations towards online crowdfunding for charity. Stud. Psychol. Behav. 2020, 18, 570–576. [Google Scholar]
  14. Sargeant, A.; West, D.C.; Jay, E. The relational determinants of nonprofit web site fundraising effectiveness: An exploratory study. Nonprofit. Manag. Leadersh. 2007, 18, 141–156. [Google Scholar] [CrossRef]
  15. Sura, S.; Ahn, J.; Lee, O. Factors influencing intention to donate via social network site (SNS): From asian’s perspective. Telemat. Inform. 2017, 34, 164–176. [Google Scholar] [CrossRef]
  16. Bin-Nashwan, S.A.; Sarea, A.; Al-Daihani, M.; Ado, A.B.; Begum, H.; Alosaimi, M.H.; Abdul-Jabbar, H.; Abdelsalam, M.K. Fundraising appeals for the covid-19 epidemic fight: A cross-country study of donor responses. Sustainability 2022, 14, 6486. [Google Scholar] [CrossRef]
  17. Briscese, G. Generous by default: A field experiment on designing defaults that align with past behaviour on charitable giving. J. Econ. Psychol. 2019, 74, 102187. [Google Scholar] [CrossRef]
  18. Ke, X. An analysis on donation-based crowdfunding with the background of charity law: Current situations, challenges and responses. J. Guizhou Univ. Financ. Econ. 2017, 35, 53–60. [Google Scholar]
  19. Shi, X.; Fan, Z.; Gao, J. Implementation evaluation and sustainable development of China’s religious charity policy. Sustainability 2022, 14, 1656. [Google Scholar] [CrossRef]
  20. Chen, X.; Yue, D. The influence of linguistic style on the fundraising performance of donation-based crowdfunding projects. Technol. Eco. 2021, 40, 111–121. [Google Scholar]
  21. Majumdar, A.; Bose, I. My words for your pizza: An analysis of persuasive narratives in online crowdfunding. Inf. Manag. 2018, 55, 781–794. [Google Scholar] [CrossRef]
  22. Zhang, F.; Xue, B.; Li, Y.; Li, H.; Liu, Q. Effect of textual features on the success of medical crowdfunding: Model development and econometric analysis from the Tencent charity platform. J. Med. Internet Res. 2021, 23, e22395. [Google Scholar] [CrossRef]
  23. Du, Y.; Jia, L. Configuration perspective and qualitative comparative analysis (QCA): A new way of management research. Manag. World. 2017, 6, 155–167. [Google Scholar] [CrossRef]
  24. Rihoux, B.; Ragin, C.C. Configurational Comparative Methods: Qualitative Comparative Analysis (QCA) and Related Techniques; Sage: Thousand Oaks, CA, USA, 2009. [Google Scholar]
  25. Smith, G.E.; Berger, P.D. The impact of direct marketing appeals on charitable marketing effectiveness. J. Acad. Mark. Sci. 1996, 24, 219–231. [Google Scholar] [CrossRef]
  26. Baumann, D.J.; Cialdini, R.B.; Kendrick, D.T. Altruism as hedonism: Helping and self-gratification as equivalent responses. J. Pers. Soc. Psychol. 1981, 40, 1039–1046. [Google Scholar] [CrossRef]
  27. Zhang, J.; Xu, X.; Liu, Q. The effect of information framework and individual emotion elicitation on college student’s donation behavior. Stud. Psychol. Behav. 2019, 17, 360–367. [Google Scholar]
  28. Fiss, P.C. A set-theoretic approach to organizational configurations. Acad. Manag. Rev. 2007, 32, 1180–1198. [Google Scholar] [CrossRef] [Green Version]
  29. Ragin, C.C.; Fiss, P.C. Net effects analysis versus configurational analysis: An empirical demonstration. In Redesigning Social Inquiry: Fuzzy Sets and Beyond; University of Chicago Press: Chicago, IL, USA, 2008; Volume 240, pp. 190–212. [Google Scholar]
  30. Ragin, C.C. The Comparative Method: Moving Beyond Qualitative and Quantitative Strategies; University of California Press: Berkeley, CA, USA, 1987. [Google Scholar]
  31. Zhang, M.; Du, Y. Qualitative comparative analysis (QCA) in management and organization research: Position, tactics, and directions. Chin. J. Manag. 2019, 16, 1312–1323. [Google Scholar]
  32. Hatfield, E.; Cacioppo, J.T.; Rapson, R.L. Emotional contagion. Curr. Dir. Psychol. Sci. 1993, 2, 96–100. [Google Scholar] [CrossRef]
  33. Fredrickson, B.L.; Branigan, C. Positive emotions broaden the scope of attention and thought-action repertoires. Cogn. Emot. 2005, 19, 313–332. [Google Scholar] [CrossRef]
  34. Lee, D.; Park, J. The relationship between a charity crowdfunding project’s contents and donors’ participation: An empirical study with deep learning methodologies. Comput. Hum. Behav. 2020, 106, 106261. [Google Scholar] [CrossRef]
  35. Gasiorowska, A.; Chaplin, L.N.; Zaleskiewicz, T.; Wygrab, S.; Vohs, K.D. Money cues increase agency and decrease prosociality among children: Early signs of market-mode behaviors. Psychol. Sci. 2016, 27, 331–344. [Google Scholar] [CrossRef]
  36. Vohs, K.D.; Mead, N.L.; Goode, M.R. The psychological consequences of money. Science 2006, 314, 1154–1156. [Google Scholar] [CrossRef] [Green Version]
  37. Li, M.; Li, J.; Tan, M.; Li, H.E.; Zhong, Y. Exposure to money modulates the processing of charitable donation: An event-related potentials study. Neurosci. Lett. 2021, 765, 136277. [Google Scholar] [CrossRef]
  38. Piff, P.K.; Stancato, D.M.; Côté, S.; Mendoza-Denton, R.; Keltner, D. Higher social class predicts increased unethical behavior. Proc. Natl. Acad. Sci. USA 2012, 109, 4086–4091. [Google Scholar] [CrossRef]
  39. Kraus, M.W.; Piff, P.K.; Mendoza-Denton, R.; Rheinschmidt, M.L.; Keltner, D. Social class, solipsism, and contextualism: How the rich are different from the poor. Psychol. Rev. 2012, 119, 546–572. [Google Scholar] [CrossRef] [Green Version]
  40. Yi, S.; Hao, F.; Liu, C. The impact of activation of money concept and preference for monetary function on altruism following ego depletion. J. Psychol. Sci. 2021, 44, 954–959. [Google Scholar] [CrossRef]
  41. Schwartz, S.H. Normative influence on altruism. In Advances in Experimental Social Psychology; Elsevier: Amsterdam, The Netherlands, 1977; Volume 10, pp. 221–279. [Google Scholar]
  42. Chen, S.; Wei, H.; Meng, L. The impact of congruency between moral appeal and social perception on charitable donation. J. Psychol. Sci. 2019, 51, 1351–1362. [Google Scholar] [CrossRef] [Green Version]
  43. Hansen, S.L.; Eisner, M.I.; Pfaller, L.; Schicktanz, S. “Are you in or are you out?!” Moral appeals to the public in organ donation poster campaigns: A multimodal and ethical analysis. Health. Commun. 2018, 33, 1020–1034. [Google Scholar] [CrossRef]
  44. Ragin, C.C. Redesigning Social Inquiry: Fuzzy Sets and Beyond; University of Chicago Press: Chicago, IL, USA, 2008. [Google Scholar]
  45. Schneider, C.Q.; Wagemann, C. Set-Theoretic Methods for the Social Sciences: A Guide to Qualitative Comparative Analysis; Cambridge University Press: Cambridge, UK, 2012. [Google Scholar]
  46. Robbie, R.W.; Billie, S. Creating meaningfulness in public service work: A qualitative comparative analysis of public and nonprofit managers’ experience of work. Am. Rev. Public Adm. 2022, 52, 122–138. [Google Scholar] [CrossRef]
  47. Tang, G.; Wang, F. What contributes to the sustainability of self-organized non-profit collaboration in disaster relief? A fuzzy-set qualitative comparative analysis. Public. Manag. Rev. 2022, 24, 466–488. [Google Scholar] [CrossRef]
  48. Liu, R.; Huo, Y.; He, J.; Zuo, D.; Qiu, Z.; Zhao, J. The effects of institution-driven entrepreneurial education in chinese universities: A qualitative comparative analysis approach. Front. Psychol. 2021, 12, 719476. [Google Scholar] [CrossRef]
  49. García-Pascual, F.; Pérez-Campos, C.; García Sánchez, J.; Soto-Rubio, A.; Aguado Berenguer, S. Models of sports management in fitness centres. Influence of sex, age and sport frequency. Linear models vs. qualitative comparative analysis. Sustainability 2021, 13, 8995. [Google Scholar] [CrossRef]
  50. Song, Y.K.; He, L.J.; Zhang, H.J.; Lan, X.Y.; Wang, H.C.; Yang, Y.Y. Survey on residents’ perceptions and usage of online fundraising platforms in nanchong city—Take “waterdrop” as an example. Int. J. China Mark. 2019, 36, 188–189. [Google Scholar] [CrossRef]
  51. Bennett, A.; Elman, C. Qualitative research: Recent developments in case study methods. Annu. Rev. Political. Sci. 2008, 9, 455–476. [Google Scholar] [CrossRef]
  52. Meer, J. Effects of the price of charitable giving: Evidence from an online crowdfunding platform. J. Econ. Behav. Organ. 2014, 103, 113–124. [Google Scholar] [CrossRef]
  53. Rich, J.A.; Corbin, T.J.; Jacoby, S.F.; Webster, J.L.; Richmond, T.S. Pathways to help-seeking among black male trauma survivors: A fuzzy set qualitative comparative analysis. J. Trauma. Stress. 2020, 33, 528–540. [Google Scholar] [CrossRef]
  54. Yang, W.; Xie, J. An analysis of factors affecting the credibility crisis of charitable organizations from the perspective of internet. Shandong. S.S. 2021, 10, 124–131. [Google Scholar] [CrossRef]
  55. Boehm, C. Moral Origins: The Evolution of Virtue, Altruism, and Shame; Basic Books: New York, NY, USA, 2012. [Google Scholar]
  56. Fiss, P.C. Building better causal theories: A fuzzy set approach to typologies in organization research. Acad. Manag. J. 2011, 54, 393–420. [Google Scholar] [CrossRef] [Green Version]
  57. Champion, V.L.; Skinner, C.S.; Menon, U.; Rawl, S.; Giesler, R.B.; Monahan, P.; Daggy, J. A breast cancer fear scale: Psychometric development. J. Health Psychol. 2004, 9, 753–762. [Google Scholar] [CrossRef] [Green Version]
  58. Grol, M.; Koster, E.H.W.; Raedt, R.D.; Bruyneel, L. Effects of positive mood on attention broadening for self-related information. Psychol. Res. 2014, 78, 566–573. [Google Scholar] [CrossRef] [Green Version]
  59. Madurapperuma, M.A.Y.D.; Kim, K.M. Sustaining business: A psychological perspective of donation behavior. Sustainability 2020, 12, 9355. [Google Scholar] [CrossRef]
  60. Zhu, Z.; Xiang, F. Research on influencing factors of fundraising effect of internet medical crowdfunding project based on panel data. Chin. J. HS. 2021, 38, 725–728. [Google Scholar]
  61. Wang, Z.; Wang, X. Trust and donation: The role of social network in donation-based crowdfunding. Manag. Wld. 2021, 37, 4–7, 34–50. [Google Scholar] [CrossRef]
  62. Kim, T.; Por, M.H.; Yang, S.B. Winning the crowd in online fundraising platforms: The roles of founder and project features. Electron. Commer. Res. Appl. 2017, 25, 86–94. [Google Scholar] [CrossRef]
  63. Mehrabian, A.; Russell, J.A. An Approach to Environment Psychology; The MIT Press: Cambridge, MA, USA, 1974. [Google Scholar]
  64. Jacoby, J. Stimulus-organism-response reconsidered: An evolutionary step in modeling (consumer) behavior. J. Consum. Psychol. 2002, 12, 51–57. [Google Scholar] [CrossRef]
Sustainability 15 01094 g001 550
Figure 1. The research model on the relationship between help-seeking message content and online charitable behavior.
Figure 1. The research model on the relationship between help-seeking message content and online charitable behavior.
Sustainability 15 01094 g001
Sustainability 15 01094 g002 550
Figure 2. The QCA method flow chart of this study.
Figure 2. The QCA method flow chart of this study.
Sustainability 15 01094 g002
Sustainability 15 01094 g003 550
Figure 3. The mechanism of how help-seeking message content affects online charitable behavior.
Figure 3. The mechanism of how help-seeking message content affects online charitable behavior.
Sustainability 15 01094 g003
Table
Table 1. The coding rules of variables.
Table 1. The coding rules of variables.
VariableCoding DimensionCoding Rules
Condition variablerational appealsname
age
home address
disease name
treatment process
positive emotionsthe ratio of positive words, such as “happy,” “safe,” “healthy,” “warm,” “reunion,” “sunshine,” and “harmony,” to the length of the message
negative emotionsthe ratio of negative words, such as “devastated,” “anxious,” “cornered,” “tormented,” “painful,” “unable to eat and sleep,” and “crying,” to the length of the message
economic conditionoccupation (e.g., farmer, peasant, or worker)
family members (including children and elderly being supported)
household income
treatment expense
moral appealsany moral discourse in messages such as “There is no mercy in illness, but there is love in life,” “May goodness be treated gently by the years, may goodness have the most beautiful reincarnation,” and “May good thoughts live on forever, may love be passed on”
Outcome variableonline charitable behaviorfundraising rate = the charitable donation amount/the target fundraising amount
Table
Table 2. Table of calibrated results for all variables.
Table 2. Table of calibrated results for all variables.
CPENERAMAECOCBCPENERAMAECOCB
10.950.990.810.51210.050.02110.250.47
20.680.02110.75122110.610.50.4
3110.610.51230.0510.610.250.28
40.740.020.810.251240.050.010.810.750.23
50.830.980.6100.99250.050.711110.19
60.780.880.810.50.99260.270.68100.750.15
710.88110.250.99270.050.56100.750.12
80.680.270.610.50.98280.630.060.400.50.09
90.690.04110.50.97290.390.56100.250.07
100.150.050.810.50.95300.050.980.600.750.06
110.90.99110.250.95310.980.03100.250.04
1211110.50.95320.160.080.800.250.03
130.990.14110.250.91330.350.780.810.250.02
140.180.05110.250.89340.970.80.6100.01
150.680.97110.250.85350.050.010.600.250.01
160.050.050.610.50.75360.050.170.800.50.01
1710.76110.50.69370.050.570.800.50.01
180.050.270.810.50.62380.050.430.610.750
1911100.50.62390.950.280.410.50
200.050.010.810.750.54400.050.050.6000
Note: C = case; PE = positive emotions; NE = negative emotions; RA = rational appeals; MA = moral appeals; EC = economic condition; OCB = online charitable behavior.
Table
Table 3. Necessity analysis results for conditions.
Table 3. Necessity analysis results for conditions.
High Level of Online Charitable BehaviorLow Level of Online Charitable Behavior
Condition variableConsistencyCoverageConsistencyCoverage
Positive emotions0.69140.69770.37470.3651
Negative emotions0.56430.58430.51840.4930
Rational appeals0.89610.55190.19770.5026
Moral appeals0.93900.66500.06100.1008
Economic condition0.51540.58400.68030.5996
Table
Table 4. Intermediate solution for sufficiency analysis.
Table 4. Intermediate solution for sufficiency analysis.
Condition VariableConfiguration 1Configuration 2Configuration 3
Positive emotion ⚫⚫
Negative emotion⊗ ⊗
Rational appeals•••
Moral appeals⚫⚫⚫
Economic condition⊗⊗
Consistency0.82970.86070.8505
Raw coverage0.32930.45790.2410
Unique coverage0.13260.26120.0444
Overall solution consistency0.8647
Overall solution coverage0.6349
Note: ⚫ or • indicates the presence of the condition, ⊗ or ⊗ indicates the absence of the condition, ⚫ or ⊗ indicates a core condition, and a blank space indicates that it does not matter if the condition is present or absent.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Liu, Y.; Cao, R.; Wang, Z. Does Help-Seeking Message Content Impact Online Charitable Behavior? A Qualitative Comparative Analysis Based on 40 Waterdrop Projects. Sustainability 2023, 15, 1094. https://doi.org/10.3390/su15021094

AMA Style

Liu Y, Cao R, Wang Z. Does Help-Seeking Message Content Impact Online Charitable Behavior? A Qualitative Comparative Analysis Based on 40 Waterdrop Projects. Sustainability. 2023; 15(2):1094. https://doi.org/10.3390/su15021094

Chicago/Turabian Style

Liu, Yanzhi, Rong Cao, and Zheng Wang. 2023. "Does Help-Seeking Message Content Impact Online Charitable Behavior? A Qualitative Comparative Analysis Based on 40 Waterdrop Projects" Sustainability 15, no. 2: 1094. https://doi.org/10.3390/su15021094

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop