Table 1

A Student Guide to Understanding the Article

 Tips, Comments, and Problem Areas

Abstract

“causal impact”: affect, influence

“implicitly”: in a subtle way, without  directly saying

“plural pronoun”: in this case, using “we” rather than using two singular pronouns such as “she and I”

Introduction

Dispositional attributions: judgments that a person’s behavior is due to the person (rather than to the situation).

Nuanced: subtly different

Causal role: the effect

Bidrectional: both ways (in the context of this article, language may affect thought and thought may affects language)

Abstract language: in the context of this article, describing a person in terms of traits rather than in terms of what they actually did or using abstract predicates (e.g., attacked) rather than concrete predicates (e.g., pushed).

Structural linguistic cues: in the context of this article, how abstract the language is.

Priming: presenting a stimulus (often to activate some ideas or knowledge associated with that stimulus)

Mediating mechanism: how (the way) a stimulus has its effect (for more about mediating mechanisms, see pages 50-51 of your text).

Study 1

Overview

Method

“incidental”: neither intentional nor important

“causal influence”: affect, influence, cause, make

The Participants section is fairly clear—although the authors should have mentioned that participants were randomly assigned to questionnaire condition.

Materials and Procedure

Fairly straightforward—although you might not be clear about the IOS. The following may help you visualize the circles used in the IOS. Participants chose one of seven pairs of circles, Approximations to two of the pairs are displayed below.

    Self               Other                                               Self  Other

Results and Discussion

Paragraph 1— The researchers combined the scores from the 6 different rating scales (the different scales are described in the first sentence of the previous paragraph) into one score by adding up each participants’ ratings for the 6 scales and dividing by 6. The researchers felt justified in combining the scores because the ratings correlated highly with each other. They used a coefficient alpha as a measure of how well scores on the scales correlated with each other. Their obtained correlation alpha (.87) is acceptably high.

Paragraph 2—

If you just follow the means (abbreviated “M”), you will understand the main point.

Univariate:  “uni” means one (a unicycle has only one wheel), “variate”  refers to variable. Because the researchers had reduced the six dependent variables to one, they could use a univariate  (one dependent variable) test.

ANOVA: analysis of variance; a general statistical test that is described in Chapter 10. If they had only been looking at the relationship between their two pronoun conditions and scores on a measure, they could have used either a t test or an ANOVA. However, they were looking at the relationship between more than one predictor on scores on their measures. Therefore, they needed to use an ANOVA instead of a t test (To understand why, see Chapter 10).

Predicted (significant) main effect of condition: Participants in the we condition rated the relationship as closer than participants in the Valerie and I condition—and this difference was reliable.

F (1, 171) = 8.75, p = .004: The p value is below .05, telling you that the observed relationship is probably reliable (see Chapters 9 and 10 for more about the meaning of statistical significance). F stands for F test, the test ANOVA uses. The first number in parentheses represents the degrees of freedom for the predictor. Specifically, it equals the number of conditions minus 1. Thus, in this case, the first number (1) indicates that there are two groups (we versus Valerie and I. In reporting the t test, people do not write out this first degree of freedom because it would always be 1 (because the t test always compares 2 conditions and 2-1 is always 1). The second number (171) represents the degrees of freedom for the error term. If the researchers had used a t test instead of an F test to analyze the effects of pronouns, the researchers would probably have gotten similar results which they might have summarized as follows:

t (173) = 2.95, p < .05.

The authors did not use a t test because the authors had two predictors (language condition and gender).

“Interaction”:  The effect of a variable depending on the level of another variable (Interactions are covered extensively in Chapter 11). In this case, a statistically significant interaction would have meant that the effect of pronouns (going from we to Valerie and I) would be different for men than for women. For example, it might have been that men were more influenced by the pronoun manipulation than men. However, in this case, it is not critical that you understand the interaction because it was not statistically significant.

Paragraph 3- similar results to those described in paragraph 2, except that the authors are trying to predict participants’ score on the IOS rather than trying to predict participants’ average score on the six other rating scales. If you are having trouble with any of the terminology (e.g., univariate ANOVA, main effect of condition, F, interact), see our paragraph 2 notes.

Paragraphs 4 and 5 summarize the main findings and conclusions.

     “Incidental”: slight, apparently unintentional

     “Explicit choices”: deliberate decisions