American college students might spend hundreds of hours a year polishing their Facebook profiles, but there is no evidence that they are more narcissistic than their counterparts a decade or two ago, according to a paper to appear in the February issue of Psychological Science.

The new study challenges a much-discussed 2007 paper that warned of a tide of crippling student narcissism. That paper’s lead author, Jean M. Twenge, an associate professor of psychology at San Diego State University, told The Chronicle last year, “We have to stop telling kids they’re special all the time.”

In their paper, Ms. Twenge and colleagues conducted a meta-analysis of dozens of small studies in which college students had taken a questionnaire known as the Narcissistic Personality Inventory. They found that average NPI scores had risen sharply since 1982.

The new paper looks at NPI scores from a narrower source: students who took the NPI in introductory psychology classes on the Berkeley and Davis campuses of the University of California in 1996, and each year from 2002 to 2007. Nearly 26,000 students in those classes answered the questionnaire; by way of comparison, the total number of students in the small studies in Ms. Twenge’s paper was around 16,000.

The authors of the new paper — Kali H. Trzesniewski, an assistant professor of psychology at the University of Western Ontario; M. Brent Donnellan, an assistant professor at Michigan State University; and Richard W. Robins, a professor of psychology at the University of California at Davis — argue that their Psych 101 sample is more representative of the broad student population than were the small pockets of students covered in Ms. Twenge’s study.

Ms. Trzesniewski and her colleagues found no evidence that NPI scores had risen since 1996. They also compared their findings to average scores that were gathered 20 years ago by the NPI’s creators. If anything, Ms. Trzesniewski and her colleagues write, scores have declined slightly since the 1980s.

Separately, the new paper looks at measures of “self-enhancement,” which is the propensity to overestimate one’s skills. Ms. Trzesniewski and her colleagues found no evidence that students are more likely to overestimate their intelligence (relative to objective measures like grades and standardized-test scores) than they were three decades ago.

Ms. Twenge will publish a response to the new paper in a coming issue of the Journal of Personality, according to an article about the dispute in today’s New York Times. —David Glenn