Civic honesty around the globe

See allHide authors and affiliations

Science  05 Jul 2019:
Vol. 365, Issue 6448, pp. 70-73
DOI: 10.1126/science.aau8712

eLetters is an online forum for ongoing peer review. Submission of eLetters are open to all. eLetters are not edited, proofread, or indexed.  Please read our Terms of Service before submitting your own eLetter.

Compose eLetter

Plain text

  • Plain text
    No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Author Information
First or given name, e.g. 'Peter'.
Your last, or family, name, e.g. 'MacMoody'.
Your email address, e.g.
Your role and/or occupation, e.g. 'Orthopedic Surgeon'.
Your organization or institution (if applicable), e.g. 'Royal Free Hospital'.
Statement of Competing Interests

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.

Vertical Tabs

  • Editor's Note

    We thank the many readers who commented on this paper and the authors for their reply.  Eletters on this paper are now closed.

    Competing Interests: None declared.
  • Reply by authors
    • Alain Cohn, Assistant Professor, University of Michigan, School of Information
    • Other Contributors:
      • Michel Maréchal, Professor, University of Zurich, Department of Economics
      • David Tannenbaum, Assistant Professor, University of Utah, Department of Management
      • Christian Zünd, PhD student, University of Zurich, Department of Economics

    We have heard from concerned Chinese readers about the results of our recent study, and here we respond to many of the questions and objections that have been raised.

    Before diving into specifics, it is important to note that cross-country comparisons were not the focus of our paper. We largely focused on how lost wallet reporting rates were affected by the amount of money inside the wallet. Cross-country comparisons were discussed only in 2-3 sentences at the end of the paper, primarily to highlight its potential use as a data set for future scholarly research. In the paper we did not single-out or discuss reporting rates for any particular country, because that was not the point of our paper.

    We emphasize our focus on explaining differences between wallets with and without money because this goal guided all methodological decisions. This general fact helps to address many of the questions and concerns raised by readers about our study design.

    Below, we respond to the most common questions or objections to our study.

    1. The only contact information available to report the missing wallet was the owner’s email address. However, many people in China communicate using WeChat rather than traditional email. Could this fact explain the relative low reporting rates in China? Why didn’t the researchers allow for other ways to report the wallet besides by email?

    A challenge of any field research such as our is to find a standardized paradigm that...

    Show More
    Competing Interests: None declared.
  • RE: Civic honesty around the globe

    The article is on controversial on the Chinese Internet.

    Some brief comments:

    - The author ignores the cultural communication traditions so leading a completely biased experimental results. It would be suggested to consider some factors of different active probability of using emails for communication. Most local Chinese hotel staff do not even own an email or they are traditionly used only for business agreement with extra authorizations. I know some similar experiments with a minor modification that allows social media name on the name cards (e.g. WeChat) achieved a completely different result in China. If you put only WeChat information on name card in some other country other than China, you may get a result that similar to China reported by the paper but happen in other countries. People may want to assist but in this case they do not know how to do that.

    - China has similar pattern with other Asian country, like Japan. People tend to put the founded losts to a lost and found area or local police office nearby. People who lose their items may found the lost by themselves.

    - The survey result has very limited causality to the paper statement. i.e. It is hard to show a directly causality between "lost return rate" to "Civic honesty".

    Competing Interests: None declared.
  • What count as honesty?
    • Xinyue Zhou, Professor, Zhejiang University
    • Other Contributors:
      • Yacheng Sun, Professor, Tsinghua University

    In an admirable attempt to investigate “civic honesty around the globe”, Cohn et al. turned in 17,000 “lost” wallets at public and private institutions in 40 countries. We conducted a similar field experiment by dropping 407 wallets (containing money and contact) in highly trafficked areas of three major Chinese university campuses [1]. In our experiment, 24% of the wallets were actively reported to the owner or turned in, which is remarkably similar to the email contact rate in China reported by Cohen et al. Among the rest, only 5% went missing, presumably taken, while 71% were left untouched. It is therefore interesting to ask: What do our results tell about the level of civic honesty in China? 24% or 95%?

    Honesty is defined as morally correct conduct [2]. In scenarios involving property, dishonesty is typically measured by the tendency of taking possession of others' property [3-5]. Apparently, not contacting the owner of a lost wallet is not equivalent to taking possession of it, as our findings suggest [1]. The receivers in Cohn et al.'s experiment might have passed the wallet to lost-and-found or just waited for the owner to pick up. Unfortunately, Cohn et al.’s data are silent about these alternative behaviors. Importantly, these alternatives are also morally correct; whether actively contacting or passively waiting for the owner is therefore non-diagnostic about whether the receiver is honest.

    The specific behavior that a receiver tends to a...

    Show More
    Competing Interests: None declared.
  • Reconsidering the methodology towards China in “Civic Honesty around the Globe”
    Competing Interests: None declared.
  • To add a few points

    1. public service organizations such as bank and museum may have different work load for different Countries. In heavey populated Counties there might not have human effort to deal with such trival matters.
    2.There are many possible reasons that wallet is not retuened, it may not reflect honst status (only when people take a lost thing for his/her own use can be defined as dishonest in this experiment regard) while instead it may reflect the way people deal with lost_found items: For example, in my university (Sichuan university), every canteen and dormitory bulding has a fixed location to store lost-found items. If some one thought he/she lost it near that building, he/she should go and search the lost store location himself/herself.
    3. If email has to be involved in this experiment, there might be a good chance that experiment object in China would use "qq" email which could be blocked or sent to the junk mailbox. (This happened to me for quite a few times. I think many people in the world do not view "qq" email as a formal mail. But I would say why not? What's wrong with this email?)
    4. Science may consider expand to a more diversed reviewer team to aviod the publiction of such biased and unscientific research.

    Competing Interests: None declared.
  • RE: Call for unbiased 'honesty' studies
    • Bo Xia, Graduate student, NYU Langone Health

    In the Report “Civic honesty around the globe” (June 20th), Cohn, et al conducted a field study of “lost-wallet report” to model the civic honesty (1). Surprisingly, the authors found a counter-intuitive result that financial temptation increases civic honesty (2). Despite the interesting results, I’m deeply concerned by the experimental design which comes with strong confounding factors. Ultimately, such biased results may propagate a fallacious and even prejudiced view of civic honesty across populations.

    The reporting approach through email as provided by the business card in the study clearly led to unreal reporting rates of the “lost wallet”. The authors failed to consider email penetration rates among people across different countries, neither did they provide alternative reporting approaches. Obviously, the countries with a low rate of email usage – regardless of having access to computers or not – will have low rates of reporting the lost wallet. For example, in China, emails are barely used by people working in the agencies where the authors conducted the experiments (3). Instead, telephone and mobile devices (and its associated social media) are much more widely used both in life and in working environments (3,4). Providing a phone number on the business card would greatly increase the reporting rates because of its availability and simplicity. Similar lack of email usage also exists in other regions, especially in developing countries (5,6).


    Show More
    Competing Interests: None declared.
  • RE: Civic honesty of China
    • Hanjun Ai, M.S., Zhejiang Sci-Tech University

    Alain Cohn, Michel André Maréchal and colleagues investigated civic honesty around the globe. I believe there was a flaw in this research. The only contact information, business cards in the wallets, just displayed the owner’s name, email address, and job title. But for many countries, email is not the usual way of contact, for example, Chinese people usually use telephones and WeChat. I believe that addressing this issue could further increase the value of their study.

    Competing Interests: None declared.
  • RE: A Non-Scientific Article Published in "Science"

    1. Does the survey consider the different traits and features of different countries and regions?
    2. Why did the assistants only presented Name and Email in the wallet? Is email the only common communication tool?
    3. Does this "experiment" understand the culture and lifestyle of the tested countries?
    4. Does the designer consulted any local people prior to conducting the survey?
    5. The fundamental premise of any global experiment is to respect local knowledge and cultural relativism, which is literally missed in this paper.
    6. Does the authors realize how the negative impacts will be on these tested countries, when they published such a biased conclusion?

    I suggest to repeal the publishing of this paper, and further scientific and thorough experiments should be done in order to make objective judgement.

    Competing Interests: None declared.
  • RE: The experiment design and methodology ought to be re-considered
    • Shixian Du, Undergraduate Student, UC Davis School of Engineering

    The experiment design itself does not seem to reflect the actual "civic honesty" of people in each country.

    First, 400 wallets per country is a fixed quota according to this experiment design. However, 400 might not reflect as well in a country like China where population are much greater than those in Europe. So statistically the result is not actually indicative and reflective.

    Second, this methodology does not take the difference of culture and social behaviors into consideration. For example, in China, few has prioritized emailing as a way of communicating with others. Many in China probably does not use email at all. Therefore, there is no hope for those people who might collect the wallet to approach you and return your wallet. Does it mean the Chinese people are "the least honest"? Probably not.

    The trick that this journal article played is that they "smartly" correlates "didn't respond with emailing" with "dishonest". Still, this correlation needs further proof from the sociologists, and by any means is, in fact, incorrect because no consideration of other factors such as the percentage of active email users is regarded. Additionally, the methodology section indeed never show an example of the email address that the experiment team has used. What if the email address is and people in China are not supposed to use gmail? What if the notes in "local language" is auto-t...

    Show More
    Competing Interests: None declared.
  • RE: EMAIL is never a common communication software for Chinese.

    EMAIL is never a common communication software for Chinese. Chinese tend to use telephone call, text and wechat or QQ. Because this it means this article have problem. It may not only happen in china, but all the list countries. The author should pull back this article and do more research, especially go to different countries culture and study what communication tool is the most common in the culture then redo the research and post again.

    Competing Interests: None declared.
  • RE: What's your problem?

    I am a Chinese college student.I only use email when receiving authentication information.If I find this wallet on campus, I'll send an email.But I don't think the selected service staff will easily use the email.

    Competing Interests: None declared.
  • RE: The experimental design is very imprecise

    In China, we use wechat or phone number to contact each other.e-mail address?Are you living in ancient times?Take a look of this:

    Competing Interests: None declared.
  • RE: This article is not scientific as expected.

    Dear Science,

    The method used in this article isn't scientific at all. It doesn't consider the difference of communication methods among different countries. In China, email isn't the most popular way of contacting other people, instead cellphone number or wechat id are the most popular ways. Please check the following video ( ). A group of expats, living in China, did an anonymous test. The difference is they left the cellphone number and wechat id in the wallet. The overall return rate of the wallets is over 73%. It is way higher than what is presented in this article.

    As the most reputable science publisher, I c hope you can check the article more carefully. It is very unfortunate to publish an article with misleading info in it.

    Your faithful reader,

    Competing Interests: None declared.
  • RE: Measurement error from respondent behavior

    As a social science researcher grown up in Chinese culture, I do have thoughts on the measurement error we might have to proxy civic honesty by the rate of returning a wallet, especially in China.

    It seems we are assuming that front desk staff would be regarded as dishonest if not returning the wallet, but it might also be some other scenarios I can list that are highly likely to happen in the experiment but relate little to measuring the honesty of the respondent:

    1. The respondent is too occupied in work to remember or to invest in returning some money of potentially insignificant amount (in fact putting $13 instead of $130 might bring completely different outcome, while I personally think the latter might bring a more convincing measure of honesty)
    2. The communication channel adds to the burden and effort to contact. In this case, e-mail. I feel comfortable to say a significant number of Chinese residents now rarely use e-mail.
    3. The wallet is headed over to the police instead.

    I can keep on listing them (and other contributors are doing it as well), though I believe the authors should have been aware of these distinctions and errors. And this is not China specific-they exist in all cultures. However, the magnitude might vary potentially a lot across. I believe these issues would need to be addressed if the article is trying to convey such a concrete message (when it quantifies and ranks countries) about a moral value to a broad set...

    Show More
    Competing Interests: None declared.
  • RE: honest work

    It is difficult to make a "scientific" commentary of the article. What is sociologically interesting is how the results certainly please a part of the world that historically self-praise by its moral superiority... I guess some historical factors account for the "likes " of the article in the European press.

    Competing Interests: None declared.
  • RE: biased experiment design

    In response to this article with biased design, there is another experiment conducted in China. The video is available on Youtube under the title "Putting the Public's Honesty to the Test in China...". Contrary to the result in this article, the reporting rate is much higher. I'm not justifying that the experiment design in the video is better or Chinese people are more honest than others. I'm questioning the experiment design itself and how this article can be published on Science!!! Please watch the video and if you are interested in this topic please leaving your comments if any. Thank you very much.

    Competing Interests: None declared.
  • RE: Biased contact method - only email?

    Before designing the method of this paper, have you done a research on how is the prevalence of email usage in each country? As far as I know, email usage is way less than phone calls or texts in China. If you add more contact methods in the wallets and do this experiment again, you might find very different result.

    Competing Interests: None declared.
  • RE: Civic honesty around the globe

    Similar test has been done again in China but the results looks different.

    It seems the results in this paper cannot be reproduced.

    Competing Interests: None declared.
  • RE: I prefer to give your wallet back by pigeons LMAO

    Hey there, I have your lost wallet, I'll send a lovely pigeon with a hand-wrote letter to you, and wish you can get it back by the end of next year. LOL.
    I cannot understand why you guys are still using physical wallets. Probably Chinese people just don't know that that thing is a wallet, which is a pocket-size case for holding papers and paper money, because they are using digital pay for years.
    If it's so important and urgent to you, why you wish to communicate through E-mail instead of phone or other instant chatting apps? No one uses E-mail for such purposes there. Do you aware that we are in 21st-century?
    But it quite makes sense to me, you guys are using a physical wallet anyways. LOVE & PEACE

    Competing Interests: None declared.
  • RE: Methodological flaws and stereotypical assumptions!

    This paper is a joke. The methodological flaws and stereotypical assumptions are so obvious that even someone not trained as a scientist can easily spot the fallacies. This makes a mockery of the credibility we associate with Science as a leading journal. Shame!

    Competing Interests: None declared.
  • RE: Methodology needs more improvements
    • Yanguo Xin, Cardiologist, West China Hospital of Sichuan University

    I just read thi paper, and there are some issues about the methodology.
    Firstly, it's about the wallets, only 400 wallets were "lost" in China, too small sample. And the wallets were picked up "by accident" by a study assisstant and handed in. However, alomost half Chinese could not speak English. How come this happens?
    Secondly, it's about the email box, according to Internet information center of China, the utilization rate in China 2018 is only about 38.1%. How Chinese can give back the lost wallets? I think different contact ways should be involved in different cultures. If the researchers used wechat (the most popular social method in China), there will be a totally inversion about the China data.
    I noticed that the authors responsed readers' question about why Japan was not tested? The authors said that Japan has a lot of small "police booths". However, there are also many Lost and Found in China. It does not make sense at all.
    Finally, I do not know how this redicular study could pass peer-review and be published on Science. I just hope the authors could make some improvements of it. Give me your responses.

    Competing Interests: None declared.
  • A few comments
    • Qin Hui, Data analyst, Emory University

    I went through the Replication data provided by the authors in the Acknowledgments, especially the part about China. I have a few comments to share:
    1. About email usage. The authors used World Bank data on "how often business firms use email to contact their customers" to control the results and found it has no big effect. The number for China is 85%, and is pretty high among the countries which have this data (for example, only ~76% for Germany). Apparently using such data to control the results wouldn't help China's ranking. Even if this data is reliable, it is not clear how it reflects the usage of email for people involved in this study in their daily life. I suggest a survey about email usage in daily communication should be conducted at the same time in institutions similar to those the experiments were performed at.

    2. According to the data, all 400 "wallets" were "turned in" in China by a single study assistant (#1, male, age 27.7) between July 7-27 2015. Each day he delivers 30-40 "wallets", each to a different institution, which I feel very intense though may not be impossible. About 55% of the recipients didn't speak English, or 45% did, which is pretty high for a country like China, though still much lower than the average among all 40 countries (~70%). According to explanation by the authors I found somewhere else, if the recipients do not speak English, the study assistant will use a cellp...

    Show More
    Competing Interests: None declared.
  • RE: My boss says it's just not a good paper

    Dear Editors,

    You must be curious as to why I used the plural form of editor here. Because I believe a fierce debate had been conducted between (among) you regarding the publishability of this paper due to the premature and unsubstantiated conclusion on the global civic honesty ranking.

    I, myself, am very surprised by how such weakly organized experimental results passed their own internal peer-review, the finicky Science reviewers' vetting, and, most importantly, the editors' quality control (QC).

    As a PhD holder myself, I fully understand that for a research project with a far-reaching conclusion to be published in Science (or any other journals), well-rounded and carefully executed experiments, conclusions strongly supported by the results, and high reproduciblilities are the pre-requisites. However, I see none in this paper.

    This research used the reporting rate of a lost 'wallet' with/without money among the 400 samples to measure the civic honesty of 40 nations. The general direction was not problematic, but there are quite a few critical questions the authors need to ask themselves before writing the manuscript, or, even before they performed the experiments.

    Are ~400 samples per country big enough to draw a conclusion on a global scale?
    America has more than 300 million people, can 400 represent them? How about China with a population of 1.4 billion?
    Why was not Japan included? Because Japan has a l...

    Show More
    Competing Interests: None declared.
  • RE: Cherry picking of data
    • Yue Zhao, Postdoctoral researcher, KU Leuven

    I just read this: when asked about why Japan was not included in the study, the third author of the paper said:

    "We originally planned to include Japan but after some initial pilot testing we realized that the country was unsuitable for methodological reasons. Japan has a lot of small "police booths" where people can return lost objects. During our pilot tests, we found that Japanese citizens would not contact the owner but instead drop them off at a nearby police booth. This feature made it virtually impossible for us to assign individual wallets to particular drop-off locations."

    This causes grave concern. Who and what determine whether a country is not suitable for the study based on "methodological reasons"? Is there a clear cut? China has PaiChuSuo (local police station) and lost-and-found places as well; why was China included but not Japan? A wallet that is not returned is a wallet not returned; why where it ended up in Japan is important, and suddenly becomes (at least seemingly) inconsequential when it comes to China? If the pilot test in Japan showed nobody returned the wallet and hence the study was canceled there for "methodological reasons", why was the study not canceled in other countries when a similar low rate of return was also observed, when other "methodological reasons" could very well be at play?

    According to the explanation offered by the author earlier, had the study include...

    Show More
    Competing Interests: None declared.
  • RE: Misunderstanding is fatal
    • Yihang Ren, underguaduate, University of Nottingham

    Most of the comments evaluated and questioned the mothod of this experiment, however I would not. The only thing I want to express is that the title 'civic honesty around the globe' can be confusing. When someone has only taken a look at the title and figure 1, it may simply consider this experiment as a process of ranking civic honesty of these nations, which is the information that most chinese media are spreading right now. Many chinese people at home or abroad are indigent about this. However that is reckless and rediculous since the honesty cannot be quantified and this whole article is actually examining the relationship between civic honesty and material incentives. I also believe that you did not want to lead to such a great misunderstanding in the first place.

    So please consider my humble suggestion, change the title into a more suitable one such as 'How can honesty be affected by material incentives?' or simply 'the relationship between civic honesty and material incentives'. Moreover, please state it in the article that 'the results may not be 100% correct to reflect the reality of civic honesty in these nations since the cultural differences always exist around the world and experimental error may influence as well.'

    Hope you can continue to work on more academic and less controversial research and articles in the future. Good luck.

    Competing Interests: None declared.
  • More problems in the methods
    • Rongfeng Cui, Ph.D. Postdoc, Max Planck Institute for Biology of Ageing

    After reading through the supplementary information, I identified further issues in the methodology that requires clarification from the authors.
    It appears that the grocery list was given in the local language, but the fictional names were given in Roman alphabet in the supplemental tables. During the experiment, what writing system was used to print these names? If the personal names were not printed in Chinese characters, for example , it is difficult to believe that test subjects would even take the business cards seriously. Author should show a photograph of the package used in the experiment from China to address concerns. A complete list of all locations should be published along with the dataset, so that the results can be independently verified.
    Second, it was mentioned that “German-speaking” research assistants were employed. In what language did they speak to the test subjects? They were asked to make a range of assessments by providing subjective scores of the test subjects. Where did the authors deposit their full dataset? Why are these scores for many countries not shown? Several important control parameters are not reported, for example, the age distribution, sex and ability to speak and understand English, whether subjects understood the situation were not made available.
    From the regression analysis presented , it was clear that the results were driven by the European countries since geographical variables of “low temperature “, “high lat...

    Show More
    Competing Interests: None declared.
  • RE: Ecological Validity in Cross-Cultural Research
    • Xiang Zhou, Assistant Professor, Purdue University

    As a cross-cultural psychology researcher, I am deeply concerned with the ecological validity of the report "civic honesty around the globe" for several methodological issues common among cross-cultural research designs.

    The lost wallet design was a paradigm that has no past research supporting its predictive power or even correlates to "civic honesty" in any countries. Cohn et al. borrowed from the classic "lost letter" design, which also had limited empirical support in developing countries. Ironically, as Cohn et al. cited (reference 17) about the WEIRD (Western, Educated, Industrialized, Rich, and Democratic) problem in research, they are creating another WEIRD study by using a WEIRD framework to understand none-WEIRD populations. This is evidenced by, for example, in Kenya, they "did not carry out data collection in the last city visited (Malindi) because the research assistant was arrested and interrogated by the military police for suspicious activity (p.4 in Supplementary)" Thus, the leap from the data to the conclusion in Cohn et al. is to assume what "civic honesty" means in each of these 40 countries by using a single indicator, i.e., returning lost wallet. It is also a WEIRD design when research assistants are "all recruited from two German-speaking universities (p. 77 in Supplementary)", thus a better conclusion to be drawn for some countries might be "how Western tourists were trusted a...

    Show More
    Competing Interests: None declared.
  • RE: It shows the arrogance of researchers

    We need to notice two words,"ACTIVE" and "E-MAIL".
    The experiments are set at only "ACTIVE" connections are counted. However, if one picked a wallet and took it to "lost and found", to the professional officers, why is it classified into "NO honesty"? However , this is the most common situations in China.
    The researchers left E-MAIL numbers in wallet, however, they do not consider DIFFERENT people in DIFFERENT countries prefer DIFFERENT connections. If I walk on a street in New York and left wallets with wechat numbers, there is no doubt that less people would connect me. More than 100 million people use wechat every day in China, but how about New York? That is same to e-mails.
    So this research is not fair and objective, and contains such arrogance. I feel sorry for that.

    Competing Interests: None declared.
  • Cultural differences may generate artifacts in response rates
    • Rongfeng Cui, Ph.D, Postdoctoral Researcher, Max Planck Institute for Biology of Ageing

    Reading through the methods, I identify several shortages in the methodology, if not properly controlled for, would explain the potential artifacts in this manuscript.
    A lost key was placed inside the package. A key in Germany has a very different value than a key in China. In Germany (and perhaps other European countries), a key to an apartment opens doors of the whole building. A key to an office opens main gates of the whole institute. If a user loses a key in Germany, it means that ALL locks in the same building need to be replaced, often amounting to tens of thousands of euros. Users are also not allowed to make copies of the key, so once lost, they will not be able to get into doors. In China, a key only opens the door to a single apartment, and users often make backup copies of it, thus a seemingly lost key would bear drastically different values (more than 100 fold) in the two cultures. This may explain why Germanic countries rank higher than other countries.
    Second, email is not commonly used in China unlike the west, especially between strangers. Wechat qq or phone number is the most commonly used method. The business card with only an email address would appear to be SPAM and suspicious, especially when the domain name that looks very unfamiliar. The very fact that the wallet contains cash also contradict the payment habit in modern day China, which are exclusively carried out by mobile payment such as Wechat Pay or Alipay. Chinese people do not hav...

    Show More
    Competing Interests: None declared.
  • RE: More cautious on reasonable comparisons leading to right conclusions.

    When reading this paper “Civic honesty around the globe”, I saw many people like me disagree with the major conclusion the authors had. The major concern arise from the method the authors applied as well as emotional disagreement when people from those countries comparing to what they experienced. The method, yes indeed has a problem, it is not unbiased as the authors never had preliminary efforts tracing the culture difference that might impact the results. For instance the email usage and the common route people found their lost in different countries. However, I am concerning on an issue not uncommonly taking place: How should we control the data collecting method when we harvesting large scale dataset? Normally we applied controls for certain parameters. We applied as much as possible, even though we still worried about outset variance that might possibly make our data "dirty". However, not all people really mentioned about how the conclusion altered if too much tight criteria applied during the process, even if they were carelessly applied. Whenever the result only reflected the truth only under certain circumstance and a general conclusion has been entitled, the paper will therefore become questionable. And I am sure that this paper is definitely not an unique case. It is a good start to emphasize on not applying certain controls that might hurt the result.

    Competing Interests: None declared.
  • RE: China is unsuitable for methodological reason.
    • Yu Li, Teacher, Beijing Jiaotong University

    The authors of this article used email as statistical method, that were not adapted to local situations. Email is not popular in China even in university. When conducting statistics and rankings, the result has a problem with its methodology.

    Competing Interests: None declared.
  • RE: regarding localised materials and ethics
    • X W, PhD Student
    • Other Contributors:
      • YY H, PhD Student

    Dear Authors,

    Thank you very much for sharing!

    Regarding the results for the first study:
    1. The measure and materials for “honesty” are inappropriate:
    a) Police station: may actively contact / Other staff: waiting (at least in China, A pilot study would help here.)
    b) Email address: definitely not the most popular contact way in China. I do hope you consider the server stability as well.
    In our less than humble opinion, it should be several convenient ways for contact, even plus visit the participants who haven’t contacted: Do they still keep the wallet for someone who would come? Hey! Someone might be still hanging there for the researcher to come back!
    c) “Can you please take care of it?” is ambiguous.
    d) We did not see the advantage of a non-local recruited research assistant.
    A better measure could be: a few hours/days later, another researcher goes back to the participants for the wallet.

    In general, we wonder to which extent this research was designed and conducted locally and how many local researchers were involved for no related information was provided in the current materials.
    a) As many other comments have already mentioned, we don’t find it acceptable to design a social experiment without understanding and taking consideration of the local cultural background, in particular, when study about MORAL issue which has long and widely recognized as a factor which almost impossible to be discussed...

    Show More
    Competing Interests: None declared.
  • RE: Civic honesty around the globe
    • dingguo zhang, Doctor, First Affiliated Hospital of Nanjing Medical University

    Recently, the Science published a paper called "Civic honesty around the globe of global citizens". Among the interesting findings, their country-specific differences study found that Switzerland scored the highest overall score, followed by other European countries, including Norway, the Netherlands, Poland and Sweden, the United States and the United Kingdom in the middle, and Chinese citizens were the least willing to report missing wallets. This makes many Chinese have to ask themselves: is our social honesty really the bottom of the world?
    From the experimental data, this result is actually inconsistent with the so-called integrity research topic. First of all, its statistical method is actually active contact with the owner, not to mention how to contact. Secondly, according to their experimental results, there are more connections between wallets with money than wallets without money, and between large wallets and small wallets. Then, this data reflects not the integrity, but whether the wallet of this value is worth the game of contacting the owner.
    In this manuscript, the author designed an experiment to test how much money temptation affects social honesty. Simply put, a dozen German college students went to 40 countries around the world to lose their wallets to see if they could be returned by the people who picked them up. If it can be sent back, it means that the people in this country have a high level of Honesty. Of course, wallets are...

    Show More
    Competing Interests: None declared.
  • RE: How do you think that Science published the “unscientific” paper?

    Science published “Civic honesty around the globe”, it causes the big storm in Chinese website. Till now (2:03 2019/6/25) zhihu web about “Civic honesty around the globe, Why is Chinese honesty ranked first from the bottom?” browsing volume reached over 5,570,841, (
    To some extent, I agree with Ann C Lawrence, Jacky Yang and other commentators. I disagree to any attack. The key question is misunderstanding the sentence “Can you please take care of it?” due to different cultures. In china it is understood as “keep it carefully and don’t lose it”. If assistants express their meanings clearly, for example “please find someone who lost the wallet”, the employee will do a good job.
    In the other word, how do you think that Science (great magazine) published the “unscientific” paper is good question to be researched.

    Competing Interests: None declared.
  • RE: Comments on Civic Honesty around the globe
    • Lu Fang, Ph.D. student, University of Pennsylvania

    Thank you for sharing the study of civic honesty around the world with us. The results surprise me that higher percentage of people return the wallets when there are more, rather than less, money. However, there is a great bias in the study, and I would like to share some of my comments about this article.

    First of all, this article should consider cultural differences in distinct countries. From the example of lost wallet shown in Fig. S1. The business card only contains name, job title, and email address. I understand that email is very popular nowadays for daily communication in the United States, but the most common contact ways in China do not include email. There are many preferred contact methods, including but not limited to, Wechat, QQ, etc. Many companies, and employees share their Wechat/ QQ/ company information in a QR code on the business card. This make contacts easier ever than before. Therefore, finding out the best way to be reached by strangers in different countries is crucial. In general, a business card should provide more than one type of contact methods to make the most successful for people to reach back to the person printed on the business card, such as telephone number, and office address.

    I understand the authors provide the proof of the email usage argument by focusing on the samples in hotels only. Many corporates have rules and regulations that employees can only use their emails to interact with the customers and suppliers for...

    Show More
    Competing Interests: None declared.
  • RE: Logical flaw

    The article has glaring logical disconnections between the data they collected and the result they drew. The authors handed out "lost wallet" with contact information and considered ONLY the recipients of said wallets who actively contacted the owner are "honest". Granted, the statement "if the person who got a hold of the wallet contacted the owner, she/he must be honest" is commonly accepted to be true. However, as a fundamental theorem of Logic, this statement says nothing about the converse, that is, the verity of the statement "the person is honest only if she/he contacted the owner whose wallet she/he got a hold of" is ill-defined. Yet this is the statement the authors used to deduct their conclusion.
    Further, it is arguable that the authors' own data indicate this lack of logical connections. if "the only reason that the person, who got a hold of the lost wallet, did not contact the owner is she/he being dishonest", which is the contrapositive of the statement used in the article, then the number of reasons that can make people dishonest must be somehow POSITIVELY correlated with the rate of people being dishonest. Between the cases studied in this article, "no money" wallets and "money" wallets, the latter has one distinctively more reason for people to be dishonest, economic incentives as the authors called. Yet, in virtually all places studied, "no money" wallets cause peop...

    Show More
    Competing Interests: None declared.
  • RE: "Honesty" is not a proper word in the title

    I am wondering how the authors understand the word "honesty" in general. Will "warm-heart" be a more accurate word for the title? It is hard to predict how the recipients handled the wallets if they didn't contact the owners. In my mind, only if the recipients took the ownership of the wallets for their personal benefits, their behavior can be judged with "honesty". What if those recipients left the wallets at Lost&Found? In many countries, L&Fs don't actively contact owners but just keep the found items safe. According to authors' metrology, they are not "honest". Is it objective? For a cheap wallet with no money and IDs or only with small amount of money, it is arguable whether there is necessity to contact the owner for people from different culture background. A sociology research without counting in the factor of culture background and behavioral patterns is defective.

    Competing Interests: None declared.
  • RE: Some comments on this paper
    • Lianxu Hao, PhD student, Helmholtz Center for Infection Research

    I don't think this is a reliable and convincing research. For this, I have a few comments and doubts regarding the results of China, because I think the author has no understanding or consideration of the population and culture difference in China:
    1. I really doubt the experiment setting, you can not equal the non-writing email behavior with civic honesty. No email response does not mean dishonesty, I do not expect someone would contact me by email if I have lost my wallet in China. There should be a three option here, keep the wallet and wait for the people coming back to find it.
    2. No background noise was taking consider in this research. So I think in areas that has a high population, lost wallets are something super common for those staff to encounter, this probably will reduce the possibility to writing email. They would rather wait for the owner coming back to them.
    3. 400 samples for China, a country which has 1.3 billion people living there? The experiment setting is also 400 samples for UAE, which has a population of no more than 10 million.
    4. The things in the name card holder, the so-called wallet, to be honest, are just useless trash if there is no money inside. I do not think this will trigger altruistic behavior at all. Maybe the staff would even think this is more a prank than a real wallet.
    For these, I recommend the Science Publication reconsider and should retrive this paper, otherwise the credit and reputation of Sc...

    Show More
    Competing Interests: None declared.
  • Experiment based on laughable poor local knowledge...what a shame

    Allow me to start with a fictional example just for the benefit of the discussion:

    On a very beautiful morning, I receive an E-mail saying that a prince from an African country would share a great fortune with me, all I need to do is just starting to reply to his Email.

    Quick Question here: How many people would really reply to this Email? How many would just delete this Email right away?

    The point here is: this Email, no matter its original intent or origin, sadly falls right into a similar pattern which is well known to some people as a SCAM, therefore if anyone would try to use these lines, this Email, to conduct experiments and hope to gain any meaningful response is clearly out of their minds. And they surprisingly lack any adequate knowledge of the local culture.

    Unfortunately, this is just the case of this "wallet experiment" took place in China. There is a very famous street scam pattern in China, which just begins with a person approach the victim and saying that "I found this wallet, can you take care of it ?" then give the victim a wallet.
    The best counter approach of the "dropped wallet" scam, which is the country first part of the lecture of almost all the parent to their children when they take their first trip alone, is that "you toss that away and stay out of it, no matter how much money you see in that" This family tip in Chinese culture is just the equivalent of "Do not take...

    Show More
    Competing Interests: None declared.
  • RE: Fatal Fault: Extreme bias of the habit of using email as the contact method

    The research method contains extreme bias by asking the finders to contact the owners through email. In some countries, email is not the prefered contact method. Meanwhile, the design of the "wallets" is so childish, and it seems like a joke for the testee. In this case, the testees are intuitively reluctant to send emails to the owner especially when they are busy working and email is not considered as the main contact method.

    Why not provide a phone number and see if the testee will call the owner by phone? It will be much more reliable.

    Anyway, I believe the experiments DO NOT support any of the claims made in this paper.

    Competing Interests: None declared.
  • RE: civic honesty around the globe

    It's more of a matter of interaction model and language. These assistants are German speaking; are those employees at the counter fluid in English/German? Do those employees keep the wallet and wait until the person to come to the counter? The experiment design should include a revisit that asks the employees how they have handled the wallet.

    Competing Interests: None declared.
  • 'One fit for all social experiment' trumps Science
    • CHIHUA LI, Doctoral student, Columbia University

    Disappointed an article with such a quality can be published in a so-called leading journal 'Science'. In my humble opinion, social experiments can be better performed according to local culture and customs. Here, I did not see that. What I see is 'Trumpism' to see the world only through someone's own eyes.

    Competing Interests: None declared.
  • RE:
    • Zheng Lu, Msc in Economics, University of Warwick

    There could be an argument in regards to the basic setup of this interesting experiment. According to the article, the outcomes are either 1, contacting which implies honesty and 2, no contacting which implies dishonesty. However, there is indeed a third possiblity that was not considered which is the facilities who received the lost wallets keep them in the Lost and Found department and wait for the people who lost it to retrieve it themselves.
    In this experiment, the third yet missing outcome were covered by outcome#2-dishonesty, which might lead to a great bias to the final result.

    Competing Interests: None declared.
  • RE: Civic honesty around the globe

    By using the same amount of money in different countries, the authors seem to have missed an important factor: The purchase power of money. A sum of $13 may seem trivial to a finder in Switzerland or Norway, but in a country like Peru or Mexico may be worth much more to a finder, even enough to make the decision to keep or return the money a matter physical survival.

    Competing Interests: None declared.
  • RE: problematic experiment methods

    The authors' proxy measure for "civic honesty" is how likely the *email address* on a business card in a *business card holder* was contacted in 100 days. This can be problematic:

    First, the usage of email addresses as a common means to contact strangers is not prevalent across the world. In all likelihood, in developing countries, people do not even attempt to use the email address provided.

    Second, despite the authors' claim that the ``wallet'' is a transparent business card holder, psychologically they likely do not evoke the same kind of response. A plastic card box may not be deemed as important as a real wallet, worthy of one's time to look through its contents and contact the owner.

    Third, as another commenter said, tested subjects not being the discoverers may not give a good measure of personal honesty.

    Competing Interests: None declared.
  • RE: Civic honesty around the globe

    I wonder if part of the effect found in this study has to do with the fact that these wallets were not "found" by the subject. Rather they had a human being, who had not pulled money from the wallet, who modeled the value of honesty, and who provided direction as to the expected behavior. People are very responsive to expectations from others - we are wired for this. Additionally, the fact that the subject had received the wallet was "known." The subject would not have felt anonymous, and we know that people often make different behavioral choices when seen than when not seen. Finally, the fact that for many subjects, they were in a specific role (worker) when approached which has additional expectations regarding their behavioral choice. There are a number of perceptual and cognitive factors here that could be considered.

    Competing Interests: None declared.
  • RE: the popularity of using email to contact others is the only reason

    It really, to some extent, has nothing to do with honesty, but definitely has something to do with the habit of using email.

    Competing Interests: None declared.

Stay Connected to Science

Navigate This Article