The Input Bias: When
Quantity Trumps Quality

Research | You step into a room with a handful of other subjects and are informed you will be watching two presentations on the heart-tingling subjects of “electronic ink” and “optical switches.” The first, the researchers explain, took over eight and a half hours to prepare; the second, a disappointing 37 minutes. Afterward you are asked to judge the quality of the presentations.

Chances are, you rated the first presentation better than the second, and it’s not because you have a fondness for electronic ink; it’s simply because you were told that it took longer to prepare: A second group of subjects was told the exact opposite about the preparation times and again gave higher ratings to the longer prep-time. What’s worse: both groups told the researchers that such information shouldn’t matter when judging the quality of the presentation. But it did.

Think about this the next time your boss asks you to take on a project.

The attempt to look busy at work (even when you’re playing Space Invaders or bidding for items on e-Bay while you should be writing an invoice report, networking clients, or pumping out project proposals) is not just fodder for a Dilbert comic-strip joke. Dr. Maurice Schweitzer G’91 GrW’93, assistant professor of operations and information management at Wharton, calls it “an attempt to invoke the input bias.” According to Schweitzer, the amount of time or money put into something influences the way people perceive its quality.

Schweitzer, along with Karen Chinander of Florida Atlantic University, published their findings on input bias in the July 2003 issue of Organizational Behavior and Human Decision Processes. What they found was a direct correlation between people’s perception of input “quantity” ñtime, money, manpowerñ and their perception of output quality when comparing two items of the same caliber.

“There’s this great billboard for Lexus that says something like, ‘35,000 people took vacations in the south of France last year; none of them was a Lexus engineer,’” Schweitzer says. “I don’t really care that they went on vacation. Is it a good car? That’s what I care about. People can misperceive what these inputs are [measuring].”

Schweitzer warns that managers should watch for workers who misrepresent their work through inflated inputs. When managers are judging othersñfor example, trying to decide on who deserves a promotionñthey should shield themselves from the input measures. On the other hand, employees and managers alike need to recognize that “people are going to judge you based in part on how hard they think you’re working, or how much time you’ve spent on something. You have to manage those impressions.”

When asked about the role of input bias in product marketing, Schweitzer is quick to reply, “I would like to promote it ethically,” but “in some cases you don’t want to make things look so easy. People care about the effort and the inputs and the amount of money you spend. The perception of those things impacts the way people judge [the outcome].”

Schweitzer, who is in the process of publishing a paper on emotions and trust judgments, got into the field of social psychology while working on his Ph.D. at Wharton in the OPIM department, which combines operations management and decision processes. “It’s a fascinating field in that we are just starting to understand how our judgment works,” he says. “We have adapted over the last hundred thousand years to do some things well, but still we have room to grow. Those limitations are becoming more apparent because we have computers,” he says. “We interact with more technology, we interact with more strangers, and we have to make rather rapid and sometimes important judgments about a lot of things.”

It’s clear he enjoys his work —including the opportunity to play mischievous tricks on his subjects. One test for the input bias involved giving people different types of tea to taste and telling them that one was brewed with more expensive machinery than the other was. High-quality teas were consistently judged by their input values (in the same way the presentations were), but some samples, spiked with salt and lime juice, were intentionally foul to check whether people think something “expensive” is good, even when it is nasty. It turns out they don’t.

“We said, ‘We need to make this tea really bad.’ It was really horrible, but it wasn’t as bad as when we first started off with even more salt,” Schweitzer recalls. “My assistant came back and he said, ‘You know this tea is really terrible.’ I said, ‘Great!’ He said, ‘No, we can’t use this. Somebody spit it back at me.’”—Patrick Brugh C’05