Academia.eduAcademia.edu

Figure from:ARTICLE Lying Words: Predicting Deception From Linguistic Styles

See full PDFdownloadDownload figure
TABLE 4: Effect Sizes and Reliability of Linguistic Predictors Across Five Studies NOTE: Numbers for each study represent Cohen’s d comparing liars and truth-tellers. Effect sizes in the negative direction mean that liars used the linguistic category at a higher rate. Mean d is a weighted mean of these effect sizes. Reliability was calculated using Cronbach’s alpha.

Table 4 Effect Sizes and Reliability of Linguistic Predictors Across Five Studies NOTE: Numbers for each study represent Cohen’s d comparing liars and truth-tellers. Effect sizes in the negative direction mean that liars used the linguistic category at a higher rate. Mean d is a weighted mean of these effect sizes. Reliability was calculated using Cronbach’s alpha.

Related Figures (4)

TABLE 1: Study Demographics would be given the dollar bill. In reality, the interrogator pretended to believe all participants and all were paid $1 for their participation. During the brief interrogation, participants were taken to another room where (non- operational) skin conductance leads were attached as part of a bogus polygraph system and a video camera was pointed at the participant. The interrogator asked the participant to (a) indicate whether they had taken any money from the book, (b) describe exactly what they had done while sitting in the room, (c) describe the contents of the room, and (d) describe again exactly what they had done from the minute they entered the room. The interrogation generally lasted less than 2 min.
TABLE 1: Study Demographics would be given the dollar bill. In reality, the interrogator pretended to believe all participants and all were paid $1 for their participation. During the brief interrogation, participants were taken to another room where (non- operational) skin conductance leads were attached as part of a bogus polygraph system and a video camera was pointed at the participant. The interrogator asked the participant to (a) indicate whether they had taken any money from the book, (b) describe exactly what they had done while sitting in the room, (c) describe the contents of the room, and (d) describe again exactly what they had done from the minute they entered the room. The interrogation generally lasted less than 2 min.
TABLE 2: LIWC Categories Used in the Present Study NOTE: LIWC = Linguistic Inquiry and Word Count. # words refers to the number of words per category in the LIWC dictionary. Mean refers to the mean percentage of usage in the present studies. Word count refers to a raw number.
TABLE 2: LIWC Categories Used in the Present Study NOTE: LIWC = Linguistic Inquiry and Word Count. # words refers to the number of words per category in the LIWC dictionary. Mean refers to the mean percentage of usage in the present studies. Word count refers to a raw number.
NOTE: LIWC = Linguistic Inquiry and Word Count. For % accuracy the three percentages listed for each equation are (a) % of liars identi fied accurately, (b) % of truth-tellers identified accurately, and (c) overall accuracy. For overall accuracy rates, *p < .05 and **p < .00] when compared to chance performance of 50%. Coefficients in the negative direction mean that liars used the category ata higher rate. Al coefficients are significant at p< .05 or better. TABLE 3: Predictors of Deception: Logistic Regression Coefficients Used in Prediction Equations
NOTE: LIWC = Linguistic Inquiry and Word Count. For % accuracy the three percentages listed for each equation are (a) % of liars identi fied accurately, (b) % of truth-tellers identified accurately, and (c) overall accuracy. For overall accuracy rates, *p < .05 and **p < .00] when compared to chance performance of 50%. Coefficients in the negative direction mean that liars used the category ata higher rate. Al coefficients are significant at p< .05 or better. TABLE 3: Predictors of Deception: Logistic Regression Coefficients Used in Prediction Equations
TABLE 5: Comparison of Human Judges’ Ratings With LIWC’s Pre- diction Equations in Three Abortion Studies NOTE: LIWC = Linguistic Inquiry and Word Count. N= 400 communi- cations. The overall hit rate was 67% for LIWC and 52% for judges; these were significantly different, z= 6.25, p< .001. LIWC performed significantly better than chance, z= 6.80, p< .001, but the judges did not, z= .80, ns. See the text for an analysis of error rates.
TABLE 5: Comparison of Human Judges’ Ratings With LIWC’s Pre- diction Equations in Three Abortion Studies NOTE: LIWC = Linguistic Inquiry and Word Count. N= 400 communi- cations. The overall hit rate was 67% for LIWC and 52% for judges; these were significantly different, z= 6.25, p< .001. LIWC performed significantly better than chance, z= 6.80, p< .001, but the judges did not, z= .80, ns. See the text for an analysis of error rates.