Have a look at this paragraph, which appears in a recently published paper by Xingsong Shi and Peter Franklin in the Asia Pacific Journal of Human Resources ("Business expatriates’ cross-cultural adaptation and their job performance", vol. 52, pp. 193–214):
Acculturation and cross-cultural adaptation have long been favoured subjects of research. Business expatriates represent one important group of sojourners, and previous research, reported on by Black and Mendenhall (1989), reveals that between 16% and 40% of expatriate managers return prematurely from their assignment, premature return being used as a measure of failure. Black, Mendenhall and Oddou (1991), Graf (2004) and Black and Mendenhall (1989) report that a failed expatriate assignment may cost organizations between $50 000 and $150 000 per return on average. Even if some doubts have been raised about the accuracy of these estimates, e.g. by Harzing (2002), the costs are no doubt high rather than low. Black and Mendenhall (1989) also point out that some further costs of expatriation failure cannot be easily calculated, such as the damage to the morale in the companies, the damage to the relationships with local business partners and governments, as well as losses in corporate reputations and business opportunities.
The first thing to notice is the age of the sources. We are being given a 25-year-old source to support the claim (made in the present tense in 2014) that between 16 and 40 per cent of expatriate assignments fail. Surely, either our knowledge or the reality, or both, has changed since then? This is even stranger in the case of costs, where we're being given a dollar figure without being told how it has been adjusted for inflation with sources ranging from 1989 to 2004. It turns out (I've looked at the sources) that the figure is reported in the 1989 Black and Mendenhall paper, which means that, here in 2014, we're being given a figure that's not just 25 years out of date, but missing 25 years of inflation. Indeed, Shi and Franklin simply misreport Graf's 2004 figure of $200,000-1.2 million. Graf's figure, however, is itself somewhat vaguely based on three sources published between 1986 and 1996, and also without being explicitly adjusted for inflation. "Doubts ... about the accuracy of these estimates", indeed!
Which brings us to the deeper problem with this paragraph. Just what exactly are those "doubts" that were raised? The paper they are referring to is Anne-Wil Harzing's absolutely eviscerating takedown of the myth of high expatriate failure rates and the sloppy scholarship that perpetuates it. Her paper is organised around twelve rules for correct referencing, showing how every one of them is broken by scholars who cite high expatriate failure rates. (Rule number 6: Do not misrepresent the content of the source. Rule 9: Do not cite out-of-date references. Rule 11: Do not try to reason away conflicting evidence.) It is not merely a critique of Black and Mendenhall's (or anyone else's) "estimate". Here's how she puts it:
The 16-40 EFR range became popular in the late 1980s and, by the late 1990s, became the most cited EFR figure. The origin of this EFR range, however, is unclear. None of the studies used to substantiate this claim actually mentions the 16-40 per cent range. Only one article (Shilling, 1993) mentions the 16-40 range without referring to other articles, but the claim is not based on empirical evidence. The Shilling article, however, does not appear to be the source of the 16-40 per cent figure either since only one other publication refers to this article. Most publications that mention the 16-40 per cent figure refer back to Black (1988), even though Black quotes a 20-40 per cent range. Since three subsequent articles by Black and co-authors mention the 16-40 per cent percentage, the 16 per cent lower boundary may well be simply a slip of the pen.
That is, after reading Harzing 2002 you are not obligated to merely note that "doubts have been raised about the accuracy" of the 16-40 per cent expatriate failure range before assuring us that "the costs are no doubt [!] high rather than low". You are obligated not to report that range because it has, literally, no factual basis, as Harzing points out:
Black’s 1988 statement that: ‘Studies [Baker and Ivancevich, 1971; Tung, 1981] have found that between 20 to 40 per cent of the expatriate managers do not successfully make the transition and return early’ (p. 277), however, fuels a whole series of attributions. […] Quite an achievement for a source article—Black (1988)—that offers as support one publication that does not include any failure rates (Baker and Ivancevich, 1971), and one publication in which only 7 per cent of the American companies have failure rates in this range (Tung, 1981).
There you have it. In 2014 researchers are reporting an "estimate" of expatriate failure rates that stem from egregiously misreading a 33-year-old study, while simultaneously misreading a paper (Harzing 2002) that shows that this very figure is the result of egregious misreading and endemic poor scholarship in the field. The title of Harzing's paper? "Are our referencing errors undermining our scholarship and credibility?"
I'm sure one day I'll find a paper that dutifully notes that I have "raised doubts" about the accuracy of Weick's story about those Hungarian soldiers in the Alps, being careful, of course, not to mention the plagiarism and, for good measure, plagiarising a couple of sentences from my paper!