When I was a student decades ago, at a ¡°good¡± Australian university, most teachers thought and taught but did little or no research. The mantra soon shifted to ¡°publish or perish¡±, prompting at least some academics to churn out books and articles by the dozen. Now, though, volume is not enough: you have to go straight to the top.

I work in two very distinct universities: one is among the Asian and global elite, the other is a modest institution that has aspirations of doing better. But academics at both are told by their leaders that it is better not to publish at all than to publish in anything less than a top-ranked journal.

In both cases, this approach is utterly wrong-headed ¨C especially in social science.

Research in physics, medicine or biology has global resonance (science is universal), but this is not so with the social sciences. A study of differences between voters in western Sydney and those in western New South Wales is not going to get a look-in at the American Political Science Review, no matter how well constructed the study or how rigorous the execution, because it would be considered parochial.

³ÉÈËVRÊÓƵ

Instead, such journals prize methodological sophistication and very high analytical ability above all else. Method triumphs over content. What this leads to is the dissection of trivia with an ever-sharper knife. I?have often seen young academics take some inconsequential data and manipulate them until the pips squeak. After the umpteenth regression, some relationship between two variables may emerge, resulting in a paper whose conclusion notes that more research needs to be done to explore this apparent connection.

When I was dean at a major US university, a job applicant presented a talk on offences, punitiveness and locations. She used dazzling statistical techniques to analyse the data, and my colleagues quizzed her on various measures and statistical tests. At the end, I?asked what policy advice might be given to state or local governments based on her research. My colleagues glared at me, and the applicant replied that she did not know. She did not get the job, but the paper was published in a top-tier journal.

³ÉÈËVRÊÓƵ

The emphasis on writing only for the top journals means that stories about how people live get told less and less often. And when they are told, they are confined either to journals aimed at practitioners or to popular outlets ¨C from both of which there is usually more to learn than from the elite journals, therefore.

Always aiming for the top also goes against the grain of a learning enterprise, in which people do an apprenticeship ¨C a PhD ¨C and progress from level to level. Junior academics should, of course, be encouraged to take on the world; but if anyone should be expected to do so, it should be only senior professors. Juniors will learn from their apprenticeship and make a valuable contribution even if their work is not published in the top journals.

But focusing only on elite journals is very bad for the morale of all thoughtful academics. The process of getting published is tortuous and, despite the use of peer review, not always particularly meritocratic.

The top journals pride themselves on rejecting the overwhelming proportion of articles they receive. Sometimes articles are rejected without being sent out for peer review. This is at least mercifully quick. What normally happens is that the rejection comes many months after submission because of the length of time it takes to find good reviewers. Sometimes it is accompanied by constructive and helpful feedback from the reviewers, but other times by cruel and disparaging comments from unscrupulous rivals. I?recently was told about a rising star in a US university who proudly told her colleague that she rejected all the articles that came her way because she did not want her competitors to get ahead. Such people are doubtless the exception rather than the rule, but there is no checking on such behaviour.

³ÉÈËVRÊÓƵ

In the unlikely event that the manuscript is ultimately accepted, it might be published online within a few weeks, but will wait in a queue for about two years ¨C sometimes more ¨C before being assigned volume and page numbers, alerting more readers to its existence and allowing it to be cited properly.

Our university leaders tell us that such citations are terribly important. But surely readership is the key metric ¨C and, moreover, readership by those beyond our peer groups. In the social sciences, you would hope that articles are read not merely by other academics but also by students and, above all, practitioners. If our work has no relevance to such people, university leaders and academics alike should be asking themselves what all this effort, expense and stress in pursuit of global excellence is really worth.

Adam Graycar is professor of public policy at Flinders University.