For this assignment, use data from W1 Assignment 3. This week, you will explore the hypothesis that a relationship exists between the misinformation effect (the type of information relayed) and the ac

Final Project

Geanine Patterson

South University Online

Behavioral Science for Statistics / PSY2008

February 11, 2018

Abstract

Chi-Square Test

Chi-Square DF P-Value

Pearson 8.23 1 0.0041

Likelihood Ratio 8.41 1 0.0037

Description of Participants

A total of 72 participants were used in this sample, 36 males and 36 females. They ranged in age from 18 years of age to 60. The sample of participants were from varied ethnic groups and each were placed into groups considered to be of different stress levels and their confidence levels in their memories were also different. Both the stress levels and confidence levels were tested. At the beginning of the project, a confederate of the researchers was placed in a room with a single participant of the study. This accomplice then fed misinformation to the party in the study on the color of the car used in the video of the bank robbery, which had been used for the research. One group of students was shown a video with a green car, the other group was shown a blue car. The students were asked again after two weeks and a third time after a month had passed.

Results

The results of this study showed eyewitness testimony to be flawed due to time from incident to trial and the levels of stress the participants of the study were under. The less stress a person was under, the more consistent were the answers given. According to the researchers, a person’s memory deteriorates over time, making it less likely that an eyewitness can remember all the details that were noticed at the time of the occurrence.

In this analysis n1 = 36, sample mean1 = 101.61, sample standard deviation 1 = 27.14,

n2 = 36, sample mean 2 = 99.19, Sample standard deviation 2 = 25.7.

The analysis was testing the correlation between the high-stress levels and low levels of confidence. In the correlation testing the r – the score is -0.707533 and the p-value <0.0001. The results show us a significant relation between stress levels and confidence levels. For example, if a businessman/woman are trying to impress a new client with a product. When high-stress levels are present with low confidence levels, They would have a less likelihood of the presentation impressing the client.

Null Hypothesis, There is no difference in the population mean recall one regarding three stress-level conditions (low, medium, and high).

Alternative Hypothesis There is a significant difference in the population mean recall one regarding three stress-level conditions (low, medium, and high). At least two groups are significantly different.

µ1 = population mean recall 1 for low stress-level

µ2 = population mean recall 1 for medium stress-level

µ3 = population mean recall 1 for high stress-level

Level of significance = 0.05

Since we have more than two groups, we will use one-way analysis of variance to test the hypothesis about equality of means

The one way ANOVA is used to test the population means recall one concerning three stress-level conditions (low, medium, and high). Population mean for recall 1 differed significantly across the three stress-levels, F (2, 69) = 18.76, p = .0001. Tukey post-hoc comparisons of three groups indicate that the high-level stress group (M = 5.208) gave significantly lesser mean recall at the time.’

We conclude that there is a significant difference in the population mean recall at Time 1 concerning three stress-level conditions (low, medium, and high). Further, from pair-wise p- values table (Multiple Comparisons), we also notice that the p-value is < 0.05 for the two pairs High-medium and high-low.

In this analysis n1 = 36, sample mean1 = 101.61, sample standard deviation 1 = 27.14,

n2 = 36, sample mean 2 = 99.19, Sample standard deviation 2 = 25.7.

Null Hypothesis (Ho): µ1 = µ2

Alternative Hypothesis (Ha): µ1 ≠ µ2

µ1 = population mean confidence levels for consistent feedback on the car color and µ2 = population mean confidence levels for inconsistent feedback on the car color.

Level of Significance = 0.05

Lower Critical value at 0.05 with n1+n2-2 = 36+36-2 = 72-2 = 70 DF is plus or minus 1.994

Reject Ho if test statistics value is smaller than the smallest critical value or if test statistics value is larger than the upper critical value.

F = (Variance of Sample 1)/ (Variance of Sample 2) = (27.14*27.14)/(25.7*25.7) = 1.11

Upper Critical value at 5% with (36,36) df = 1.75

Since the test statistic values are less than the upper critical value, we can conclude the variance is equal for both the samples. And we will use independent sample t-test for an equal variance to test the original test statistics.

The test statistics value is 0.388470098 which is smaller than 1.994 and more significant than -1.994, so we will not reject the null hypothesis. There is not much difference in the population mean or confidence levels for consistent feedback on car color as well as inconsistent feedback on car color.

Discussion

The project the researchers was to test the accuracy of eyewitness testimony for trial purposes. The scientists involved in this study showed that eyewitness testimony is unreliable, especially over time. The scientists showed that eyewitness testimony could also be influenced by outside information (misinformation). By injecting a different thought into someone’s memory by suggesting a different set of circumstances, the researchers were able to influence what that particular person remembered, for instance, the color of a car.

Taking into consideration the results of the research conducted by the scientists, and about the hypothesis and the literary review, I achieved the same results the group conducting the study reached, eyewitness testimony is unreliable over a period.

References

Projects weeks 7,8,9 Behavioral science Retrieved February 15,,2018