Oppenheimer, 2006

Background

• writing experts recommend simplicity
• students intentionally add complexity and larger words to their work
• use of complex words is correlated with presumed evaluations of intellect Research question: How do people perceive the author based on the use of large works?

Experiment 1

1. does increasing the complexity of text succeed in making the author appear more intelligent?
2. to what extent does the success of this strategy depend on the quality of the original, simpler writing?
3. if the strategy is unsuccessful, is the failure of the strategy due to loss of fluency?

Procedure

• Used 6 grad admissions papers
• Subsitutsed original nouns and verbs words with their longest applicable thesaurus entries
• Exposed original and new text to Standford students and asked them to rate the
• asked students to rate the difficulty of the text and whether to accept the student

Findings

"Complex texts were less likely than clear texts to lead to acceptance decisions in a simulated admissions review... this trend was found regardless of the quality of the original essay. Complexity neither disguised the shortcomings of poor essays nor enhanced the appeal of high-quality essays."

I wonder how much of this has to do with how the complexity was added to the texts. They note this is a limitation.

Experment 2:

Compare two passages of identical content and different complexity passages.

Procedure

• used two translations of Descartes passage with differing complexity
• same procedure as the first experiment, but asks students to rate intelligence and difficulty
• half were told the passage was written by Descartes

Findings

• replicated the first experiment
• only difference is that students rate intelligence higher when they knew it was Descartes (but the effect was still there)

Experiment 3

Replicate expt 1 but simplified texts

Procedure

• used dissertation extracts instead of grad application essays

Findings

Simplicity wins

"It is the use of overly complex words—not the word replacement process—that leads to decreased ratings of intelligence."

Experment 4

Compared differences in font to effect fluence without changing the content/writing. Found that the more difficult fonts were associated with lower fluency and lower perceived intelligence even though the subjects attributed the font choice to the experimenters

If, as Experiments 1–4 suggest, fluency is the driving factor behind these effects, then one ought to be able to reverse the direction of the effect by making people aware that the source of the low fluency is irrelevant to judgment.

Experiment 5

Replication of experiment 4 but using low toner vs. normal toner when printing the texts (instead of using the font differences) produced the opposite effect because the participants associated with the experimenters and not the author (although the same was true for the font differences.)

"As predicted by the fluency account, when an obvious source for the lack of fluency is present, people discount that lack of fluency when making their judgment. They do so to such an extent that they end up biasing their judgment in the opposite direction!"

Reflections

One problem I have with this study is that it was done with elite students, who may be more adept at assessing whether a text could be reasonably simplified than typical college students or the general population. Writing quality in universes is often lamented (I know my undergraduate writing was low quality), but this might not be true at Stanford. I would like to see this replicated with different populations before calling this robust.

Oppenheimer frames this experiment as having to do with students increasing the complicity of their work for class. Often professors have limited time to review papers. (I was told to spend 5 minutes on teaching essays when I TAed at tufts.) I wonder whether this effect would hold in, say, essays for a large lecture where TAs are expected to grade 50 essays (on top of all their other work). Part of me thinks that simplicity would help the students even more in the scenario, but alternatively, when given limited time, TAs may index on things like perceived complexity of ideas which may be related to the perceived complexity of language.