It has been a long time since I have done a study. 1988, I think. Designed one or run one. A long time since I have written one, and I know I have made many errors here.
I have been telling my students that words matter. Words create perception and they can be used to create bias, emotion, action or inaction. We study appeals to pathos, logos, ethos, kairos. Loaded language and logical fallacies. But I often sense they do not believe me. So I thought I would put them in the middle of their own proof.
The result was many open eyes and one student who insisted he should be filming me as a TED talk.
The results are below.
Design
This study is designed to see if using a “charged” term, non-academic diction, can change perception of external events. Such language can be used to create bias or emotional states and it was my desire to demonstrate this to English honors and Advance Placement English Language and Composition classes. If the hypothesis is correct, this can demonstrate how “charged” terms can be used to control the overall responses of populations.
I hypothesis that using terms with a “positive charge” will increase perception of speed in a filmed vehicular accident.
Population
Three classes of tenth grade honors English students were tested, with populations of 18, 19 and 21. All classes were studying the same curriculum and in the same program at the same location in their curriculum and instructed with the same materials, methods and instructor.
Material
Each class was shown a five second film of a vehicular accident or a motorcycle striking a car that had just pulled out of a parking spot, as filled from a helmet camera. The film showed the motorcycle increasing in speed, with the sound of the engine extant, and striking the broadside of the car. It was made obvious, in the film, the rider was not hurt appreciably hurt, and there were no signs of injury in the film.
Method
Each class was asked to estimate the speed of the collision and to write the number, in miles per hour, on a note, but each class was asked using a slightly differently worded query. The control group was asked the question in academic diction devoid of purposefully charged language. A second group was asked the same question with a word replacement or a neutral for a word with a “positive charge.” The third group had a query with two words carrying a “positive charge.”
- What was the speed of the vehicle when the accident occurred?
- What was the speed of the vehicle when it smashed into the other?
- How fast was the vehicle when it smashed into the other one?
The notes were collected and the data compiled for mean, median and mode as well as lowest and highest outliers.
Population 1
27 mph average speed estimate
30 median
30 mode
Lowest outlier 4 mph. Highest Outlier 53 mph.
Population 2
33 mph average speed estimate
30 median
30 mode
Lowest outlier 12 mph. Highest outlier 55 mph.
Population 3
38 mph average speed estimate
35 median
35 – 40 split for the mode, with four estimates for each
Lowest outlier 18 mph. Highest outlier was 80 mph.
Results
The language with the least emotional charge, the academic diction, resulted in the lowest perceived mean speed as well as the lowest outliers.
The language with one added “charged” word increased the mean perceived speed 22.22% 33 mph over the control group
The median and mode did not shift but the lowest and highest perceived speed increased by 200% and 3.78% respectively over the control group.
The language with two “charged” words increased the mean perceived speed by 40.74% to 38 mph over the control.
The median increased 16.67% to 35 mph and the mode was split evenly between 35 and 40 mph. Using the mean of this mode to calculate percentage, the mode increased 25% over the control. Seemingly most telling is the increase in the outliers. The lowest perceived speed increased from 4 mph to 18 mph (350%) and 80 for the highest (50.94%) over the control group.
It is clear using charged words increased perceived speed.
This can be extrapolated to other areas, such as crowd size, levels of violence, impending danger and many other real world events.
Interpretation
This demonstrates several things. Language can be leading/loaded even if language does not appear to be. Academic diction has the lowest “charge,” and this supports the need to teach students to be write in an academic fashion. It also supports the need to instruct them to understand the importance of diction, so they can recognize language which appears to create logos when it is really designed to create pathos, thus allowing students to notice subtle manipulations in language meant to create emotional responses to skew perception and/or drive opinion. Further, it demonstrates the need for careful word choice with high semantic value to decrease linguistic indeterminacy.
Replication and Refinement
In replicating this study, I would select a population corrected for gender and academic level to assure the populations were homogeneous. Further, I would add a 4th group with a variable “negatively charged” term to see if the perception of the estimated speed in such a group would be lower than the control.
In further refinement, I would like to test to see if changing the charged adverb (fast, slowly, quickly) or the verb (smashed, collided, hit) have differing magnitudes of affect.
Discussion
We are aware that journalism can look objective but, upon examination, we find leading words and loaded language hiding in the sentences. This can have an effect on how we perceive an event. The word “mob” used for an assemblage instead of group can, and does, affect how people perceive the assemblage and this carries over to the perception of the individuals within the assemblage.
While I understand, in this test, the outliers skew the data, and it is possible the outliers should be taken into account when calculations are made, the outliers are of interest in themselves. Both ends of the outliers rose with inclusion of the charged words. The outlier at the higher end is of particular interest as it is the outliers in a society that cause the most dramatic and concentrated change and cause the most trauma as well (terrorism, murder, mass shootings) and if a small inclusion of a charged word can create a large increase in the emotional response of the top outliers, this is worth noting.
While we cannot combat this in every instance, we can begin to educate students to be aware it exists and to be on the lookout for the use of such language. Words with a positive charge can be used to excite/increase bias and bring activity when coupled with a call to action. Words with a negative charge can dampen responses and reduce activity. We see this in political rhetoric as well as in sales, and we are seeing it increasingly in social media and fringe news sources.
It is possible that educating children to recognise and not accept the charging of language may help reduce the effect of this.
Adam Byrn Tritt 