The Sample Size Effect is a very complicated subject to introduce into the field of psychology. The sample size effect is usually spoken of in the context of statistics. Usually meaning the number of participants, with this the larger the sample size the more likely you are to get accurate or precise results than if you have a small number. Not much research has been done with the sample size effect and psychology but Betsch and Kauffman (2009) was about the only study we could find that spoke about the Sample size effect in relation to psychology. In their paper The Origins of the Sample Size Effect in Explicit Evaluative Judgment (Betsch & Kauffman,2009) they concluded that “Evaluative judgments should become less extreme if the variability of the information sample increases.” You can see how this relates to statistical definition. We decided to connect this to social psychology by seeing how this idea of the sample size effect affected trust in an internet based informational source. We were interested in seeing if we used a relatively negative stigmatized internet source, Wikipedia, which has been said to be unreliable if we could change the perceived reliability (used to indicate trust in our experiment) of the participant if we varied the sample size. So, if they were told that a high percentage of their peers—fellow ivy leaguers, both students and professors— used this site and that it was reported as one of the top three websites used by these people then the participants would report trusting the information given to them from this site as more reliable and a lower trust rating if given a lower percentage. In our experiment we used an article on stem cell research taken from Wikipedia. Our diffuse question is how do we decide whether to trust an information source? And the specific question we wanted to answer with our experiment, how does other people’s perception of an information source, influence our evaluative judgment of credibility in that source? We wanted to look at how undergraduate students can be influenced to either trust or not to trust the information presented on Wikipedia. As stated before previous research has shown that the Sample-Size Effect and Perceived Reliability are factors that are involved in evaluative judgments (Kaufmann & Betsch, 09). Schaller (’90) found that even though people have more confidence drawing inferences if there are larger sample sizes there are times in which only looking at statistics can hide crucial “third variables” which would lead to different inferences. This is relevant because our results could be skewed because of the fact that people choose to trust what they read or the source more so just because of the percentages given or trust them less because of lower percentages and this is exactly what we hope to find. Finally, we found research on Internet forewarning effects (Leon, ’03) that demonstrated that people who are forewarned about possible deception, are less persuaded by the subsequent information presented to them, than those who did not receive this forewarning. Using this previous research we narrowed our scope to look at how participants’ trust in information from Wikipedia can be influenced when presented information that only a small sample of their peers use this site or that a large sample of these people use it. In support of our thoughts, in an article titled “Trust in Online Advice. Social Science Computer Review” Briggs concluded that “In a world fraught with decisions, huge numbers of people are turning to the Internet for advice and guidance with the result that online information is now exerting strong societal and personal influences”(Briggs et al, 2002).
We hypothesize there will be a sample-size effect, meaning that participants will rate a high or low trust in an information sample from Wikipedia depending on a large or small sample-size. We also hypothesize that perceived reliability will positively correlate with sample size. Participants will be asked questions that are intended to check any personal biases they may have for the site and their trust levels of internet sources and the internet itself as a manipulation check in order to examine possible situational constraints.
Save your time!
We can take care of your essay
- Proper editing and formatting
- Free revision, title page, and bibliography
- Flexible prices and money-back guarantee
Methods
Our study has one Independent Variable, the forewarning description of Wikipedia, with two levels, a report that a large number of people use Wikipedia, and a report that a small number of people use Wikipedia. Our Dependent Variable is a questionnaire measuring participants’ level of trust in the article after reading the warning. Subjects will rate the target information (stem cell article from Wikipedia) on 5 5-point scales, related to level of trust in the article they have just read and the forewarning presented prior. Scores will range from 5 (low trust) to 25(high trust). Subjects will also be given a manipulation check comprised of 3 questions that will test for retention of content from the forewarning message. If knowledge of how many individuals use an information source (sample size) affects trust in this information then a description that a large number of people use Wikipedia, should yield higher trust in information from this source. If knowledge of how many individuals use an information source (sample size) affects trust in this information then a description that a small number of people use Wikipedia, should yield lower trust in information from this source.
Participants: In condition one (large sample size) there were 21 participants, 14 female, 7 males, and the average age 22. In condition two (small sample size) there were 19 participants, 11 females, 8 males, and the average age was about 23.
Procedures:
- Step 1: Participants will read a brief description of Wikipedia
- Step 2: Participants will read an article on stem cell research taken from Wikipedia
- Step 3: Participants will complete a questionnaire with 5 questions to measure their level of trust in the information they have just read
- Step 4: Participants will complete a 3 question manipulation check to measure their beliefs about Wikipedia prior to reading the description, and to see if they actually read the material.
In Step one, participants either read one of two descriptions. In condition one where the participant read a large sample size description this is what they read, “In a recent study of Ivy League Universities, nearly 82% of students and professors report using Wikipedia for background information. Wikipedia is now the Web's third-most-popular news and information source, beating the sites of CNN and Yahoo News, according to Nielsen NetRatings” In condition two where they read a description with a small sample size they read, “In a recent study of Ivy League Universities, only 12% of students and professors report using Wikipedia for background information. However, CNN, Yahoo News, and other Internet sites beat Wikipedia as the Web’s three most popular news and information sources, according to Nielsen NetRatings.” In step two all participants were then instructed to read the following article taken from Wikipedia: “Stem Cell Controversy”
“There has been much debate in regards to the legality, and morality, of embryonic stem cell research. However, regardless of the moral implications there are numerous benefits and potential for breakthroughs in this area of medicine. Embryonic stem cells make up a significant proportion of a developing embryo, while adult stem cells exist as minor populations within a mature individual (e.g. in every 1,000 cells of the bone marrow, only 1 will be a usable stem cell). Thus, embryonic stem cells are likely to be easier to isolate and grow ex vivo than adult stem cells. They divide more rapidly than adult stem cells, potentially making it easier to generate large numbers of cells for therapeutic means. In contrast, adult stem cell might not divide fast enough to offer immediate treatment. Embryonic stem cells have greater plasticity, potentially allowing them to treat a wider range of diseases. Adult stem cells from the patient's own body might not be effective in treatment of genetic disorders. Allergenic embryonic stem cell transplantation (i.e. from a healthy donor) may be more practical in these cases than gene therapy of a patient's own cell.”In step three they were asked questions such as this one: then instructed to read the following article:
In step three participants were asked questions and asked to rate them such as this one:
- I would reference this article in conversation with my peers about the stem cell controversy
(not at all) 1 2 3 4 5 (extremely)
In step four, the participants were asked questions and to rate them such as this one:
- Did the description tell you that many Ivy league students and professors trusted the information found on Wikipedia?
(not at all) 1 2 3 4 5 (extremely)
Results
Here we have the averages rating per the first 5 questions for each condition. The large sample size ended up being 3.26 with a standard deviation of 0.66, while the small sample size condition resulted a 3.19 and a standard deviation of 0.71. Here we see that there is not a significant difference in the values and even with this small variability there is also a p-value of 0.72. Here the average trust rating is recorded on the y-axis.
Overall Average Trust
However an interesting finding was found in questions six and seven. Six asked if the description told them that many Ivy League students and professors used Wikipedia for information, while question seven asked if those same students and professors trusted the information. The participants correctly identified the correct sample size, they rated toward the end expressing that there were many of their peers who used the source if they were in condition one which had the large sample size and rated on the lower range if they were in condition two which had the smaller sample size. But in question seven when asked how much they trusted the source they all ranked it on the lower end, expressing that they did not trust the information. Here in the graph the average rating is recorded on the y-axis.
Manipulation Check
Discussion
Implicit attitudes of the source created more variability in our results. Our assumption that Wikipedia has a negative stigma was consistent across groups, despite the fact that students use the site. For example, professors often tell students not to use Wikipedia as a source, despite growing studies that suggest it’s has increased in reliability. Wikipedia has a negative stereotype. As evidence we seen this in our own results we found that most of the participants have used or were familiar with Wikipedia, however they would not use it as a citation. We also feel that another possible implicit attitude may come from how the participant felt about the article topic itself, stem cell research. Some people have negative attitudes towards stem cell research and this may have affected their trust. Consistent with another study in the same paper by Betsch & Kauffman (2009) stating that evaluative judgments based on samples of larger size but higher variability are seemingly similar to those based on samples of smaller size and low variability. This could be one reason why there was seemingly no difference between the two groups. Possible solutions to our flaws may be to use a more neutral topic for them to read in an article form. In order to further limit implicit attitudes we could use an unknown source maybe, instead of a largely controversial one such as Wikipedia. The Sample Size Effect can be brought into the world of social psychology. Hopefully our experiment can be a beginning for research on the sample size effect and perceived trust in informational sources.
Stuck on your essay?
