Cell Phone’s vs Drunk driving

I chose to look at Mythbuster’s attempt to see if talking on a cell phone while driving was as dangerous as drunk driving.  How they chose to do this is have the participants answer questions on their cell phone while driving.  This was followed by both participants drinking and taking a sobriety test to confirm that they were in fact drunk then drove the same course.  The test itself followed the scientific method by presenting the hypothesis and using methods to test and eventually confirm their hypothesis.  I personally believe that the drunk side of the test was well done because they best way to test the danger of drunk driving is to have a drunk driver in a controlled environment.  In regards to the cellphone portion of the test, I do not believe that this was highly practical because in most cell phone conversations are not done in this manner.  I believe the best solution to a cell phone conversation is rather than asking someone to repeat exactly what was said or answer a question, actually carry on a normal conversation with someone to better simulate the experience of someone driving while talking on a cell phone.  I do believe overall that this was well done because the experiment shows there is a large difference between driving drunk and driving on a cell phone although I wonder; if the scenarios were in a more practical setting such as regular conversations, would there be as large of a gap between driving drunk and driving while on a cell phone?

Research Methods- Gentlemen Prefer Blondes?

The basis of science is of course the scientific method. Since we are studying it in psychology, we were assigned to watch a “Mythbusters” and critique their methods. So, since I am a blonde and have been told my hair color is preferred by many, I was curious and picked that one. Overall, I am not surprised that there was no preference in the hair colors, since everyone has their own preference and many care more about other things than hair color. Even though the test came out how I thought it would, there were still a few issues. The first of them being how obvious the wigs were, which could have led to suspicion from the guys about them, and could have lowered the ladies’ likability if they are noticed. The other issue is that the women looked and acted fairly different so the men could’ve had a preference for something other than the hair that could boost numbers. Otherwise, the experiment was set up well, one of the better ideas being the speed dating thing with the equal numbers of hair colors and then having them all switch was a good touch. Also, getting more men in also was well planned because it allowed for a bigger sample, which allows for more empirical data, heightening the chance that their results are good.

Research Methods

In the TV show Mythbusters, episode titled “Are Women Better Than Men at Reading Emotions?”, I am observing and critiquing the methods that are being used to test this particular myth. It is being assumed that women are better at acknowledging and reading someones emotions than men are. This experiment had tested this theory by ushering in both male and females to look at pictures of just the Mythbuster crews’ eyes and taking their best shot of what emotion they are expressing before reveling the rest of their face (and the answer). It is not mentioned in the short clip if these participants are random, or if they were chosen specifically for this experiment. If these were all fans of the Mythbusters TV show, there could have been some that recognized facial expressions of their favorite person on the cast more so than others were able to. This also goes hand-in-hand with if the participant were close friends with any of the cast, too; usually we are able to recognize our friends’ facial expressions better than strangers or people that we do not know as well. This could be a flaw in their data collection due to a skew of the variety of participants. There is a good selection of both men and women viewing the slideshow; they also are all being presented in the same room, under the same conditions to observe the pictures being presented. Yet, as a viewer of the clip, it is uncertain how many comments the crew initiates towards the participant as they are looking through the photographs. Encouragement, discouragement, criticism, noises, or no verbal communication at all may be effecting participants differently, as well as their way of thinking as they work through all of the pictures. These are factors that could change the way the person sitting in front of the screen continues through the test. At the end of the clip we do learn that there is a variety of wrong and right answers between both men and women, so it is not prevalent that women are better than men at recognizing emotions. However, it is apparent that the women’s answers are much more rapid in response and seems to be more instinctive. This is a tidbit that we can take from the mini-myth being tested, yet we are not necessarily able to rely on it directly considering we are not sure of all the outside components that we do not know about.

Is It Dangerous To Drive In Heels?

While watching the Mythbusters short clip on, “Is It Dangerous to Drive in Heels?”, I found a few things to be quite problematic with the setup of the experiment and how it was conducted.  There were one or two positive things, such as the different types of shoes provided, but everything else regarding where and how they drove only one type of car was a different experiment all together!  Their driving course was put together in a way where there were enough quick turns and stops, but it wouldn’t prepare a woman going into work in early morning traffic while trying to maneuver around other drivers, or even considered driver experience, or the time of day.  Even though more than one pair of shoes were used in the experiment, I found it a bit disappointing that they didn’t include more types of heel sizes and fashions.  There are definitely all types of high heels that have higher platform bottoms and thinner, spikier heels than the shoes used in the example.  All in all, I wasn’t necessarily blown away by the research methods they used, and so I think they could have done more to make this experiment extensive and lengthy.

Hypothetical Psychology Test

Resarch Question: Does the use of stress balls decrease the stress experienced by an individual in a given situation?

Hypothesis: Stress levels for an individual will not show signs of decrease when using a stress ball to relive stress in a high-stress enviroment.


  1. Ask for volunteers around AC Campus\Sherman to participate in an academic reseach study.
  2. Randomly assign individuals to be in Group A (using the stress ball) and Group B (not using the stress ball).
  3. Place Group A and Group B in separate rooms with the same stress-inducing atmosphere.
  4. Record blood pressures of indiviuals prior to the start of the experiment.
  5. Have Group A and Group B, be in a stress-inducing enviroment for 10 minutes.
  6. Instruct Group A to reguraly squeeze a stress ball during the 10 minute time period.
  7. Instruct Group B to remain as still as possible during the 10 minute time period.
  8. Record blood pressures of individuals at the conclusion of the experiment.
  9. Compare average blood pressure of Group A and average blood pressure of Group B to determine a possible correlation between stress relief and using a stress ball.

Is Yawning Contagious


According to the TV show Mythbusters, yawning is in fact contagious due to stimulus that individuals acquire from group /social environments. This particular experiment tested the group level of analysis in Psychology by trying to control the act of yawning by placing individuals in a setting that was composed of outside influences.

A key method used in the experiment was to describe, explain, predict, and control. Also known as the goals of psychology, the testers aimed to describe how yawning was involuntary and the stimulus created from it invoked in others the act of yawning. They also explained that yawning can happen on an individual level but when done so around others it can provoke the same behavior. With fifty people tested, the researchers then predicted that yawning can be an involuntary contagious behavior. To test the possible theory of the contagiousness of yawning the experiment was controlled by a seeded yawn that was tested to discover if it would increase the behavior of yawning among a group of isolated but grouped individuals.

I would like to argue that the evidence found was in fact not supportive of their confirmation.  According to their findings the people who got no stimulus yawned 25% of the time while the people who received stimulus yawned 29% of the time. There’s undoubtedly an increase in the amount of yawning, however, it is not a significant increase. The myth busters confirmed that  yawning was contagious but from my standpoint this experiment does not support that theory in that it only shows a 4% increase in the number of yawning. However, to counter my argument a 4% increase in the amount of yawning could in fact be significant enough to confirm the theory of yawning being a contagious behavior.

Would anyone like to counter my argument?

Mythbusters first impression

Today I watched a Mythbusters video that outlined an experiment about weather or not women get bigger tips if their breast size is larger. The experiment was done by having the same woman work in a coffee shop for three days of the week. In order to control variables, she wore the same outfit for each day that she worked at the coffee shop. Using this method the team was able to test weather or not changing the woman’s breast size had any effect on how much the customers would tip. This method of experimentation had some key strengths and weaknesses. The strengths of this experiment being that the Mythbusters team was able to collect empirical data on unwitting participants, and that the Mythbusters team used the same person and outfit for the coffee shop employee. One weakness I noticed was that the test was on three different days. Having the test on three days could mean that the amount of people subjected to the test could vary, and thus the amount of tips could change based on the number of people rather than the intended variable of breast size. Another weakness was that the various customers could have varying amounts of cash to tip with. For example, a customer intending to tip the woman ,due to her breast size, might not be able to because of a lack of funds. Overall I would say that the experiment had a good amount of effort placed in its design, but with some key oversights.

Do Beer Goggles Really Exist??


Do beer goggles really exist?

The team of MythBusters are ready to chug some beers to test out the idea if the people around you become more attractive the more a person drinks… Adam, Jamie and Kari underwent a series of tests to see exactly that. These tests were to rank a set of people when they were sober, buzzed, and drunk. After each time they were given a score telling them whether or not the hypothesis was true.

Thinking about this in the eyes of the scientific method, a four part method: make an observation, develop a hypothesis, test the hypothesis, and finally build a theory. The MythBusters, firstly, made an observation that there are more male female interactions when people has some form of alcohol in their system. Secondly, they developed the hypothesis that people around you become more attractive the more someone drinks. Next, the needed to test the hypothesis, they did that by ranking a set of people while they were sober, buzzed, and drunk. The higher the score they received, that meant they saw the set of people more attractive. The lower the score they received, that meant they saw the set of people less attractive. Lastly, building a theory, coming up with a conclusion from all of the tests conducted if the hypothesis was true or not.

With every hypothesis, the tests have their own strengths and weaknesses. In this scenario, some strengths I saw included; having an intermediate test between sober and drunk, this gave them a chance to see if attractiveness is a gradual process or a sharp jump. Testing both boys and girls, giving them an idea if this hypothesis is true or false for both genders. Lastly, the tests gave them a general understanding whether or not this hypothesis is true or not. Weaknesses I noticed were, this experiment was not a real world example, meaning every time they went back to rank people there were different people shown. If this were a real world setting, you would see the same people, lets say at a bar and you’d gradually start to see them more attractive. Another would be, there were not enough people to conduct this experiment, the more the people the better the results would’ve been and the results would be more conclusive. Lastly, what if  different types of alcohol affected people in different ways. If this experiment was conducted again with changes I would propose to conduct the experiment in a real life setting, such as in a bar. This will allow the person to see similar faces and see if over time the affects of alcohol made that person more attractive. Another thing I would propose would be to add more people, with more the people in the experiment the outcome would be of a greater sample, so it could be generalized over a population. Not having enough people in an experiment makes the result hard to justify because the data could be highly varied so the average possibly wouldn’t make sense. Lastly, being able to test the experiment with other types of alcohol to see if attractiveness increases or decreases would be interesting because all people are different and not everyone gets affected the same way with beer.

After one long hangover, the MythBusters saw that overall there was a slight increase in attractiveness averaged over the three of them. Sure there were some up’s and down’s in the scores but the general trend was that as more alcohol was consumed the more a person thought everyone else was attractive.



Week 2 First Impression Prompts – Research Methods

Hand writing on a notebook

Regardless of which prompt you choose, please use the Tag “Research Methods” on your post. Here are the prompts for this week:

Option 1

For this week’s discussion, I want you to design a research study about a topic you find interesting in psychology. You can choose any topic you would like as long as you relate it to something from our textbook. In your post make sure to do the following:

  • List your research question
  • State your hypothesis (what you think the outcome would be and why)
  • Describe your procedure (what you would have participants do, how you would recruit participants)

Make your research idea something feasible that you could actually do as a student researcher. In other words, assume you have a fairly small budget and a limited amount of time. If you’re a psychology major or minor, you will eventually take Research Methods and have to conduct a research project, so this is great practice to start thinking about what you might want to do.

Option 2

Mythbusters is a popular TV show on the Discovery Channel which tests popular ideas using scientific methods.  Select one of the mini-myths (short clips from the show) below and critique the methods used to test the myth. Remember, critique means list the strengths and the weaknesses. For each weakness, discuss why it is a problem and suggest a solution.

I look forward to seeing what you write!

Header image: CC by Flickr user Caitlinator
Share Via: FacebooktwitterredditpinterestlinkedintumblrmailFacebooktwitterredditpinterestlinkedintumblrmail

Reaping What You Sow

There’s a commonly held belief of yawning being contagious. This served as the observation for a MythBusters MiniMyth , in which the team seeks out the answer to this old wive’s tale through scientific research; however, is it credible?

First off, they failed to develop a hypothesis–an essential component to any experiment because it offers the foundation for the actual gathering of data. After all, how are you supposed to test something you have no guess to? I get they were going forward with an open mind (probably because of the show’s formatting), but it would have been much better to simply say “We think this will happen.” Hence, there would be something to confirm or deny.

The actual body of the experiment, on the other hand, was well done. An independent variable (the yawn stimulus or “seeded yawn” planted by Kari) and a dependent variable (the number of yawns resulting) were established. After, a trailer with three waiting rooms was built, offering a sterile environment to rule out any outside influences. This worked towards isolating the variables so the team could more confidently discount any other possible reasons for a yawn or lack of. Rooms 1 and 2 served as the experimental group and were exposed to the yawn stimulus before entering the room whereas Room 3 was unseeded and served as the control group. An experimental group is the participant population in which the independent variable is introduced. By contrast, the control group is the participant population where the independent variable is excluded. A hidden video camera was set up in each room with Tori monitoring and keeping count of each person’s yawn. This was recorded alongside their room number and later analyzed to find the percentage of individuals who yawned without the stimulus introduced and with the stimulus introduced. The findings tallied up to be 25% yawned without the seeding yawn, and 29% with the seeded yawn, revealing a 4% difference. Conclusively, it was decided this percentage meant yawning is contagious, and the myth was labeled “Confirmed.”

This provides me with two additional issues to discuss.

For starters, the selection of participants was not random because the participants all were found from an online add calling for extras. Although they did not state the purpose of their work, it narrows the potential data pool to actors only, threatening the integrity of the experiment due to a potential sampling bias. Sampling biases can lead to skewed data, hindering the accuracy of the results.

Second, the MythBusters were too quick to confirm yawns being contagious. Experiments cannot confirm anything, only provide additional information to support or deny a hypothesis. The closest one comes to a fact in science is a law, which still leaves room for error, and includes things as omnipresent as gravity. Additionally, their experiment would have to be tested repeatedly and garner the same results at a significant level. Assumptions in science can lead to assumptions in the general population because, if this experiment and its researchers say it’s true, then it must be. While yawning being contagious isn’t necessarily dangerous, the idea still comes across: as a precaution, ever experiment’s results should be issued with a limitations section. Just in case.