Running Head: TASTE PREFERENCE IN THREE TYPES OF CHIPS

Abstract

Attempts have been made to overcome methodological errors in past taste test experiments. Students and faculty members at Lewis & Clark College made up thirty participants in a taste test experiment involving Pringles potato chips. Participants were asked to rate their preference on regular, fat free (olestra), and 1/3 fat Pringles chips in a double blind, within subject experiment. The data showed that there was a significant difference with a prefernce for the Pringles regular potato chip just as hypothesized. Despite the level of significance, the results were thrown out due to an error that occured during the data collection process of the experiment.

Taste Preference Among Three Types of Pringles Potato Chips

There have been many studies in the past that have targeted the human sense of taste in relation to certain types of food. Some studies tested a person's ability to identify the product while others were interested only in the simple taste preference one holds for a food over another. In both of these types of experiments, there have been methodological difficulties preventing any accurate conclusions to be drawn.

In an experiment performed on one's ability to identify different brands of cola through taste, some of these previous methodological errors were diminished by using new methods. Using these new methods, they were able to prove that certain brands of cola could be identified on the basis of taste (Thumin, 1962).

In our experiment we also use alternative methods of control to minimize methodological difficulties. These methods reduce bias and include such practices as conducting a within-subject design, a double blind procedure and counterbalancing.

The purpose of our experiment is to determine what type of Pringles Sour Cream and Onion potato chip people prefer most and if there is a significant difference among the taste preferences. The choices are between regular, 1/3 less fat and fat free (olestra) chips. We hypothesize that people prefer the regular chips over the low fat and fat free choices. An observation of a current social trend in reducing the amount of fat in a product for health reasons, leads to the question of taste being compensated for. We want to see if this is true in relation to Pringles and that if people prefer regular Pringles over the other two.

This report analyzing the experiment is formatted to the Publication Manual of the American Psychology Association (APA, 1994).

Methods

Students and faculty members of Lewis & Clark College who were eating or exiting the check out line in the colleges' dining area, the Trail Room, had the opportunity to participate in out taste test. We approached these people to recruit them for our study and people who stopped by our booth out of curiosity were encouraged to participate as well. There were 30 participants overall and we did not discriminate against any specific group due to their gender, age, ethnicity, or occupation nor did we target any specific group. Basically whoever showed an interest was able to participate. Due to the time of day of our experiment (10 AM to 11AM), a large percentage of our participants were women who worked at the college in the business office who were on their break. All participants were assigned to all experimental conditions and all completed the experiment. The rewards for participating were the chips they tasted at no cost to them.

Three types of Sour Cream & Onion Pringles were used in this experiment; regular, 1/3 fat, and no fat (olestra). Thirty boat-shaped dishes were used to hold one of each of the three Pringle types. Plastic cups and water were provided for the participants to clear their palate in between each chip tasting. There were also pens provided for the subjects to sign the consent form and to mark their preference rating.

The independent variables in this experiment were the three types of chips; regular, fat free and 1/3 fat. The dependent variable was the preference rating based on the Likart scale of 1 to 7. Our experiment was a within subject design so all participants received all three experimental conditions. As part of the within subject design, we used counterbalancing as a control to reduce bias and randomized the order in which they receive the treatment. The Latin square design was used in the assignment of the order of the treatment. The within subject design was used in this experiment over the between subject design because people have different taste preferences and by having them taste each chip, they could asses their preference according to their subjective taste scale. This reduces biases. To further reduce biases a double blind procedure was used. To control any extraneous variables participants drank water in between each chip tasting so the previous chip would not interfere with the current chip being consumed.

Before conducting the experiment, we took the 30 dishes and separated them into six groups. Using the Latin square design, we wrote on the bottom of each dish a sequence of three letters each one letter to correspond with one of the three types of chip later to be assigned. The sequences followed such as ABC, ACB, BAC, BCA, CBA, CAB. In the Trail Room, two of us decided which chip to assign to which letter and arranged them so on the trays. The other two experimenters were unaware of the order and set the dishes out on the table for the participants to take one at random. Thirty people who were in the Trail Room at the time of our experiment had the opportunity to participate in our taste test experiment. When they agreed to participate in our experiment, first they read over the consent form and signed it. Next we had them read a paper informing them of the nature of the experiment. It stated:

Hello! Thank you for participating in our taste test experiment. The chips are placed in these trays in order. Please eat the chips from left to right, ranking them on our taste preference scale (1 being yuck and 7 being yum). Make sure to take a drink of water and make your ranking between each chip. Thanks again and enjoy!

Following these instructions the participants performed the experiment writing their rated score for each chip on the dish in the same spot where they had just eaten the chip. After the experiment was completed, we collected these dishes and by looking through the dishes against a light source, we read the scores that corresponded to the letter. We processed this data and ran tests.

Results

After collecting the data from the experiment and performing a one-way repeated measure analysis of variance (ANOVA), we found that there was a significant difference in taste preference of the Regular Sour Cream and Onion Pringles over the other two. The mean rating for the regular chips was 5.033. The mean for the 1/3 fat was 4.400. The mean for the fat free chips was 4.267. The test results revealed a significant difference with F(2,30)=.0247, p<.05.

Discussion

The present data supports the hypothesis that the regular Pringles are more preferred than the other two. Unfortunately a major error occurred in the process of collecting data which completely discredits any of these results. In the process of assigning chips to their place in the trays the two experiments approached it quite differently. One placed the chip right on top of the letter corresponding to the chip. The other one, noticed that the order was reversed once the tray was flipped over and thought the data was going to be collected not by looking through the dish but rather by reading the data and then flipping the other side to read the other half of data. In this confusion it was impossible to distinguish between the two methods of data collection and so half of the data that was collected is false. As to which half is false is a mystery.

This test should be replicated to obtain accurate data. If it were replicated there could be other control procedures to reduce further methodological problems and biases. One thing that could have been reduced was the visual cues. There were obvious physical differences in each type of Pringle, whether it be color, or the amount of sour cream and onion mix appearing on the surface. These differences in appearances could have biases some of the participants' perceptions of the chips. Also our sample wasn't completely random. Not every student at Lewis & Clark College had an equal opportunity to participate in our experiment. Those who participated did so by coincidentally being around the Trail Room at the time of our experiment. Also we had a large percentage of middle-aged women who were on break from the business office. This percentage is not an accurate representation of the Lewis & Clark population regarding age and gender. As a result of the major error that occurred in the acquisition of data, we are unable to draw any conclusions from this experiment; but if one were to replicate this experiment with some reductions in biases and more control we would be able to accurately determine if there is a taste preference of regular Pringles.

References

Publication manual of the american psychological association (4th ed.). (1994). Washington, DC: American Psychology Association.

Thumin, F. (1962). Identification of cola beverages. Journal of Applied Psychology, 46(5). 358-360.