Author Archives: mjs5842

Random Chance and DiMaggio’s Streak

2 Replies

In my last post, I looked at how random chance plays a role in making us believe players can “get in the zone” and go on “hot streaks” when in reality, their probabilities of future success never changes. When I was reading through one of my sources for that post, I came across something that made me think of a concept Andrew brought up in class on Tuesday and I thought I would share it here:

Steven Strogatz, a professor of applied math at Cornell, set up a study to see just how unlikely Joe DiMaggio’s record 56-game hitting streak (which has been described as just about impossible). 
Strogatz used historical baseball statistics to create a computer simulation of the entire history of baseball. He ran the simulation 10,000 times and recorded the longest game hit-streak. 
Length of Record Hitting Streaks
As you can see here, DiMaggio’s streak isn’t nearly as unlikely as we may think. The observed streak length varied from 39 to 109 games. All in all, in 42% of his simulated universes, DiMaggio’s streak was at least matched. This is similar to Andrew’s car crash example. Individually, the chances of any one of us being injured or killed in a bad car crash is very small. But over the next three years, it is likely that one of the people in our classroom will have. Similarly, the chance of any one baseball player achieving a hit in 56 consecutive games is astronomically small (there’s a reason why it’s the record!). However, in the long history of baseball, it is actually fairly common for a player to go on a lucky streak of that length. In another 200 years, we will probably see another DiMaggio-like streak. Just don’t hold your breath for it to come from your favorite player next season.
I don’t know enough about “Monte Carlo” simulations to do a proper critical analysis, but I am skeptical that you can capture every aspect of a baseball season in a computer. There are effects about how good the opposing pitcher you face is on any given day, your ability of hitting certain pitches, ballpark size, and the human element of knowing you are nearing a record and “choking” under the pressure. So call me a little skeptical of the exact results he achieved, but it is still a fun exercise in statistics.
So what do you think? Are streaks like Joe’s less impressive in your eyes now?

Does “The Zone” Exist? – Free Throws

Leave a reply

We’ve all seen games like it – no matter what the defense does, a player on the opposing team just can’t be stopped. He makes every shot, every catch, or strikes out every batter. Sports, with a history full of superstition, often chalks up “hot streaks” like this to a player being “in the zone”. But does that zone really exist? Can we measure it using science?

Since we most frequently come across “the zone” hypothesis in professional sports journalism/commentary, the second analysis in this study is most interesting to me. 
The researchers looked at every field goal members of the Philadelphia 76ers took in the 1980-81 season, and separated the shots into six categories ones taken after 1/2/3 previous “hit” streaks, and 1/2/3 “miss” streaks. If the “hot hand” hypothesis held true, we would expect to seeing the chance of hitting the next streak to increase with a hit streak. In other words, the chance of making a shot would be greater after a hit than a miss on the previous shot. The probability of a hit after making two in a row would be greater than the probability of hitting after only making one in a row, and so on.
The statistics told a very different story. The weighted means for each category suggest that going on a hot streak actually makes it slightly less likely that you will make the next shot. (This seems to support basic statistics – the players were “due for” a miss to balance the hot streak and take them back to their baseline average). For eight of the nine players studied, there was not a statistically significant correlation between the outcome of previous shots and the outcome of the present shot (to put it in terms of Science 200, the p value > .05. There is a chance greater than 5% that the results were due purely to chance). 
The correlation was only statistically significant for one player, Daryl Dawkins. He experienced a correlation of -.142 that was significant at the p<.01 level, meaning that there is less than 1% chance that the result was due to chance. The statistics seem to indicate that he actually gets worse at shooting free throws as he goes on a hot streak. (Perhaps he starts to overthink what is making his shots go, throwing him off his game).
In order to control for the variable nature of athletics (a “hot” player might start getting double-teamed by the defense, lowering his probability of success on the next shot) and make sure that the 76ers weren’t just a statistical outlier, they replicated this study with the Boston Celtics in the most standardized area in the game – the free throw line. The results held. There was no statistically significant relationship between the probability of success on a free throw and the result of the shot prior to it. According to the study, any “streaks” we see are solely due to chance.
ScienceBlogs.com has a funny anecdote and about how the 76ers responded to this finding about their “hot hands”:
“The 76ers were shocked by the evidence. Andrew Toney, the shooting guard, was particularly hard to convince: he was sure that he was a streaky shooter, and went through distinct “hot” and “cold” periods. (Toney is still regarded as a great clutch player. Charles Barkley has called him “one of the best kept secrets in the history of the NBA.”) But the statistics told a different story. During the regular season, Tooney made 46 percent of all of his shots. After hitting three shots in a row-a sure sign that he was now “in the zone”-Tooney’s field goal percentage dropped to 34 percent. When Tooney thought he was “hot,” he was actually freezing cold. And when he thought he was cold, he was just getting warmed up: after missing three shots in a row, Tooney made 52 percent of his shots, which was significantly higher than his normal average.”
 
This is a good look at the battle science has against the firmly-held beliefs of populations. As an athlete, it is easily to convince yourself that you are on a hot streak and just can’t be stopped, or alternatively, blame your failures on a cold run. I believe that sports fans are also guilty of confirmation bias in assigning labels like “clutch shooter” to players – they see every last second shot made as evidence supporting the label and ignore evidence to the contrary. Here’s what perhaps the greatest basketball player ever had to say about that:
“I’ve missed more than 9000 shots in my career. I’ve lost almost 300 games. 26 times, I’ve been trusted to take the game winning shot and missed. I’ve failed over and over and over again in my life. And that is why I succeed.” – Michael Jordan


Of course, this study is merely correlational and doesn’t necessarily establish any kind of causal link or mechanism. It would be interesting to see whether any physiological changes occur in an athlete’s body during a hot streak in competition that give him an advantage. But that kind of study is difficult to imagine, and could very well change the outcome of the event studied (imagine Lebron James dribbling down the court with sensors attached all over his body, tracking the amount of adrenaline coursing through his body at any given time. With that in mind, It is extremely easy to be superstitious and irrational when dealing with a passionate area such as athletics, which makes the scientific study of it that much more interesting. In my next post, I’ll take a look at how the concept of random chance could make one of the most unbelievable sports records in history just a little less impressive…

If Homophobes are Afraid of Gays, They Sure Have a Funny Way of Showing It…

Leave a reply

A few posts down, another student blogged about pupil dilation as an arousal response. He noted that scientists chose to study pupil dilation rather than more “obvious” arousal indicators i.e. erection size, because they believed it was more purely involuntary. That brought to mind another study I had heard about, that I thought I’d share with everybody…

Psychologically speaking, a phobia is driven by anxiety is response to a stimulus. In 1996, a group of researchers set out to test the psychological theory that homophobia is driven by an anxiety over repressed homosexual urges and the fear that you may be gay yourself.
Defining Homophobia: the researchers defined homophobia as an emotional response including fear, anxiety, anger, discomfort, or aversion in response to interaction with a homosexual. This is a significant difference from the cultural use of “homophobe” to define anyone with a moral objection or intellectual argument against homosexuality – which they define as “homonegativism”. 
The Sample: Volunteers from The University of Georgia’s psychology department filled out two surveys – The Kinsey Scale and the Index of Homophobia. The Kinsey Scale is a chart that views sexual orientation as a spectrum with solely heterosexual arousal/experience on one end and purely homosexual on the other. Only participants who rated themselves a “1” (exclusively heterosexual) in terms of both arousal and experience were studied in the experiment. That pool took the Index of Homophobia, a survey which grades respondents on how uncomfortable/dreadful they feel in situations with a homosexual. From their responses, subjects were divided into a “non-homophobic” group (score 0-50 on the survey) and a “homophobic” group (score 51-100).
The Experiment: Each group was shown 3 four minute porn clips; one heterosexual, one lesbian, and one gay. Their arousal responses were measured by a “penile plethysmograph”, which tracks changes in circumference. Additionally, participants were asked to rate themselves on arousal (how “turned on” they were) and erection (“how hard” they got, for lack of a better phrase). Participants were given time to return to a base level of arousal after each clip and the order of the videos was randomly assigned to control for any order effects.
The Results: Arousal levels were similar across groups for the heterosexual and lesbian videos. Both groups showed lower levels of arousal for the gay video, but the homophobic group was significantly more affected, with a circumference change over twice as large as the non-homophobic group. The statistical analysis of the subjective ratings were interesting as well. The subjective ratings correlated really well from the arousal that was measured, indicating that the two scores were measuring the same event. But controlling for the subjective score, homophobic men seemed to have higher levels of measured arousal. This lead the researchers to hypothesize that the homophobic group under-reported their arousal, giving some credence to the idea that homophobes are in denial or repressive of homosexual urges.  
Critical Analysis: Thinking through the lens of the Science 200, this study seems really well-designed, but there are some things to think about.

  • Self-Selection: Could some of the self-identified “Kinsey 1s” lied about their sexual history? (Then again, this would probably fit under their definition of homophobia) Could some respondents have self-reported less/more homophobia on the second survey due to societal pressure to confirm to the values of Georgia? Would the results be the same in other areas of the world? Would homophobia be driven by different factors? 
  • Causes of arousal: Each of the videos included foreplay, oral sex, and penetrative sex. The arousal to the heterosexual movie increased at a steady rate. The lesbian film had a quick initial arousal followed by a leveling out. (Perhaps the foreplay/kissing was most arousing) The gay video seemed level at the beginning and end, with an uptick in the middle. (Perhaps there is something about seeing a man receive oral sex that stimulates an arousal response in men who score highly in homophobia ratings).
  • Relevancy: Would these results hold true for homophobic women. Obviously, we wouldn’t expect them to have a response to gay men as we would the men to have for lesbian women. And we would need to find another way of objective measuring arousal. And account for whether women are as easily visually stimulated.
But other than those questions, this study seemed pretty comprehensive to me. What do you think?

Don’t Let Your Sons Grow Up to be Cowboys: Country Music Causes Suicide?!

3 Replies

Steven Stack (Wayne State) and Jim Gundlach (Auburn) we’re interested in the interplay between music and society and set out to study whether country music exposure had any effect on suicide rates. Their proposed mechanism for this was through the change in psychological mood that music can lead to. Their hypothesis was that the themes of country music reinforced a suicidal mood among the “country fan” subculture.  Content Analyses of the country genre has found a high prevalence of themes that have been linked to suicide, such as divorce and jealousy. Additionally, they found that country music often encourages alcohol abuse as a normal or even necessary way of dealing with your problems. A previous study had linked exposure to country music and higher alcohol consumption. Frequent drinking seems to be linked to higher suicide risk, as implied in Wasserman’s analysis of the effect of prohibition. 

In the study, their regression analysis found that metropolitan areas with more frequent airtime had higher suicide rates, regardless of how southern the city was, divorce rate, gun availability or the poverty level. They interpreted their data to show that country music won’t drive somebody to suicide, but it could potentially put suicidal members of the subculture at even greater risk. But before you start deleting all traces of Tim McGraw from your iTunes, there are some important questions to be asked. 
1) The null hypothesis was only rejected for the white suicide rate and not black. Why? If themes of the music put listeners at risk, shouldn’t it put them all at risk? Can we just assume that African-Americans are less influenced by country music or just listen to it less than their white counterparts in the same city? Even in the south?
2) Could reverse causality be in play here? That is: rather than country music exposure causing depression/suicide risk, could depressed people seek out country music, leading to a higher demand and therefore playing time in areas with higher suicide rates?
 3) This study was only done in metropolitan areas, holding constant the variables {country music, poverty, southern region, divorce, gun availability}. Could there be some third variable that is causing the increased suicide risk that is specific to cities? Are the prevalent causes of urban suicide the same as they are in rural areas (where you could make the argument for country music being an even stronger cultural force).
4) Relevance today. Are the driving forces of suicide in the culture of 2013 the same as they were in the less technologically advanced world of 1992, the time of the survey? Is the country music landscape the same today? Maybe there is a subtle theme shift precipitated by pop-country artists (Taylor Swift, etc.), or the members of the country fan subculture themselves.
5) How were the content analyses done? I couldn’t locate the study cited in the paper at PSU, but they make a mention of a study of 1,400 hit songs. I would like to know how “hits” were determined. Even if country music “caused” suicide, maybe only niche songs were guilty. There is no way to know, and an experimental study would clearly be unethical (attempting to induce suicidal thoughts in subjects).
If you ask me, it’s a good idea to avoid country music despite your feelings on the study. But in all seriousness, I thought this paper was a fun one to read as a non-science major, and a good subject to practice critical analysis and skepticism on. 

Are Gay Couples Worse at Parenting?

2 Replies


There is little doubt that the treatment of the LGBT community is one of the largest civil rights movements in our culture today. Frank discussions are being had across the world about what kind of marriage and adoption rights gay and lesbian couples should have. One of the leading arguments against these rights is some variation of the “think of the children!” argument. And with such monumental court cases and legislation taking place so rapidly these days, that argument is worth critically examining. If scientists and sociologists can find that holding all else constant, children of gay couples are somehow “worse off” than children of heterosexual couples, we would be foolish to place children in harms way. However, if the null hypothesis (there is no difference) holds true, anti-equality activists don’t have scientifically-backed leg to stand on. Let’s look at the data:


Benjamin Seigel, a professor at Boston University’s School of Medicine, notes that there has not been a “Holy Grail” study done on the subject – a randomized, controlled trial. The difficulty behind this is fairly obvious; it would be unethical to randomly tell sets of parents which children they were “allowed” to adopt. An additional complication would be the inability to make the study “blind”, the children would always know whether or not they were raised by a gay couple. However, Dr. Seigel claims that “we’re never going to get the perfect science, but what you have right now is good-enough science. The data we have right now are good enough to know what’s good for kids”.


The major basis for his position is The National Longitudinal Lesbian Family Study. Now 27 years old, the study follows the life of 78 artificially inseminated children of lesbian couples. Among the findings of the study to this point have been:


  • The children of gay and lesbian parents are no more likely than their peers to identify as LGBT themselves.

  • In a confidential survey, none of the children reported being abused by their parents

  • The children, most recently interviewed at age 17, are relatively high-achieving in school, have close friends, and describe their parents as good role models.


In critically analyzing this study, I only had a few concerns. Dr. Nanette Gartrell, the lead researcher of the study, recruited participants in the study at women’s bookstores and events in major cities. This could potentially skew the findings (is the parenting of a avid reading urban couple the same as a random couple in the backwoods of Montana?). This also shines a light on the potential for third variables, as we always discuss in class. For example, it is probably reasonable to think that these urban couples may be wealthier than the average American parents, which could lead to a better environment for the children. Also, their choice to artificially inseminate eliminates the risk of unintended pregnancy – a large amount of family planning and reflection had to go into the decision to raise a child (obviously not the case for every child of a heterosexual couple). Additionally, the responses themselves might be subject to an unconscious bias. Although the children know their responses are confidential, there is a risk that they could artificially raise the scores on the subjective parts of the study out of reluctance to make their parents or themselves look bad (how they assess their parents are role models, disclosing their sexual orientation, etc.). But all in all, the study seems to be about as comprehensive as possible and fails to reject the null hypothesis that orientation has no effect on parenting outcomes.


Dr. Seigel acknowledges that a few studies have suggested that children of LGBT couples are in fact negatively affected by their parents, but he believes those studies should be dismissed due to fallacies much like the ones we discuss in class.


An Australian study, “Children in three contexts: family, education and social development”, surveyed teachers about the performance of students in three categories of families – heterosexual married, heterosexual cohabiting, and gay/lesbian cohabiting. The study found mixed results, with the children of gay couples faring worse than their classmates in a few categories. However, Dr. Seigel has a few issues with the findings.


  • First of all, gay marriage was not legal in Australia at the time, so based on the definition of the research, all gay and lesbian couples were cohabiting. Marriage could potentially be an outside variable that contributes to the success of children, in which case the research would be making a strong argument in favor of legalizing gay marriage at the risk of harming the children of gay couples.

  • There is reason to believe that gay families were stigmatized at the time of the research. This could act as a third variable, negatively affecting school performance and setting the students up for failure.

  • The majority of children in gay households were placed in the family after a recent divorce. That instability and stress from their family history could be the driving force behind negative performance, rather than their parents’ sexual orientation.


One final, and particularly troubling, study is the Regnerus Study. The paper by a University of Texas sociology professor found that children of gay couples we’re destined for a lifetime of negative outcomes, including welfare, unemployment, and even sexual abuse. However, the real shocker in this paper is its apparent disregard for ethical methodology.


  • The study sampled based on convenience rather than a random, representative pool the findings seem to suggest.

  • The data is questionable – some respondents claimed to have literally hundreds of sexual partners in the previous week.

  • Rather than ask for demographic data on whether their parents were in a gay relationship, the study asked whether the respondents’ thought their parents had had sex, or been in a relationship with a member of the same sex.

  • Regnerus grouped children of same-sex couples and those whose parents had suffered from a same-sex affair together.

  • The entire study only contained two children of a same sex couple – nowhere near the amount that would be needed to support the claims the study makes for the whole LGBT community.


The study was heavily criticized by the social sciences community and many researchers questioned Regnerus’ motivation in publishing the paper. One researcher quipped, ” He has been disgraced. All of the prominent people in the field know what he did and why he did it. And most of them know that he knew better. Some of them think that he’s also stupid and an ideologue. I know better. I know that he’s a smart guy and that he did this on purpose, and that it was bad, and that it was substandard.” The paper has sparked a debate on the role of religion (Regnerus is an Evangelical Christian) and politics (he also appears at events with prominent anti-equality activists across America) in research, believing that his objectivity was compromised in an attempt to influence this summer’s DOMA and Proposition 8 Supreme Court cases.


But back to gay parenting – I am inclined to agree with Dr. Seigel. Although none of the current research is ideal, the stronger studies appear to support the null hypothesis. Although there is nothing to say “gayness” makes parents any better than their peers, there doesn’t appear to be any reason to believe the opposite is true either.

Discussion: Do you know any studies to the contrary? Should we be concerned about bias in research done by strongly political or religious researchers?

Creatures of Australia: Why Andrew is Lucky To Be From New Zealand

2 Replies

Australia is home to some really scary animals. Of course, thanks to “The Crocodile Hunter” (Krikey, Mate!), alligators get all the bad press. But in reality, some of the deadliest critters are much smaller and harder to recognize. Here’s a short summary:
1. The Stinging Stonefish (Synanceia verrucosa)
This fish kills by pain. Literally. It rarely happens, but when a stonefish is stepped on by a human, “It produces such mind-blowing agony that the body goes into shock and the person dies,” according to a venom researcher at the University of Queensland. And by the way, it is called the Stonefish for a reason – it camouflages itself against rocks. So watch your step!

2. The Blue-Ringed Octopus (Hapalochlaena fasciata) 
This octopus has one of the most toxic venoms in the world, and will shut down your respiratory system and kill you in minutes with a single sting. Thankfully, it gives you a little more warning than our friend, the Stonefish. The octopus lights up in electric blue and yellow when it feels threatened.
3. Taipan (Oxyuranus microlepidotus)
Of course, Australia is also home to the most venomous snake in the world – the taipan. Mice feed on snakes. The taipan avenges the death of every other snake on the planet by feeding on mice. One snake is said to carry enough venom to kill several humans, but their location in the abandoned deserts of Queensland means that humans are safe – for now.

4. Box Jellyfish (Cubozoa)
Continuing our theme of small animals with enough venom to kill a family – the Box Jellyfish! Australians tend to take vinegar to the beach with them, because it will stop the jellyfish from injecting more venom when they try to take the tentacles off of them.
Just kidding, koalas are adorable.

If Your Friends Jumped Off a Bridge, Would You Do It Too?

5 Replies

Well…yes.

As it turns out, our brains are hard-wired for peer pressure. Scientists have found that the reward (striatum) and social reasoning (medial prefrontal cortex) centers of our brains activate much more when we engage in risky behaviors with peers as compared to the same behavior in a solitary environment. The effect is especially pronounced when we win – participants who won a lottery in the study were more likely to engage in even riskier betting behavior in future rounds. Some scientists believe this reaction has roots in evolution; in primitive times, the rewards for being at the top of the social totem pole had even greater perks (Namely, tons of food and sex. Can you really blame our ancestors?).
Everybody knows the negative influence peer pressure has on teens with regards to destructive behaviors (drug use, etc.), but its effect can go much deeper than that.
Quick – which line on the right matches the card on the left?
asch.png
Easy, right? Well if you were part of Asch’s famous conformity experiment in the 1950s, there’s a  75% chance that you would answer a similar question incorrectly at least once in 18 tries. The reason? He put actors in the experiment with you to intentionally give the wrong answer. Not wanting to be embarrassed by answering differently than the rest of the group, participants would often be “tricked” into believing the wrong answer – even for a question this blatantly obvious!
But not all hope is lost! Some social marketers see the potential for good in peer pressure. For instance, it tends to make people more generous (how many times did people give you money on canning weekends just because the people in front of them did?), reduce terrorism, and increase participation in “green” initiatives.
So what do you think? Is peer pressure good? Bad? Both? Neither? Let’s start the discussion in the comments below!

Introduction Post: Mark Shultz

2 Replies

Thumbnail image for Thumbnail image for Sheldon.jpg

Hello Everybody!,
My name is Mark Shultz, and I am a senior marketing major in the Smeal College of Business with minors in theatre and economics. On campus, my major activities are Blue Band (look for me on gamedays – I’m the first one out of the tunnel!) and Penn State Thespians.
There are two major reasons why I am taking this course. First of all, it sounds much less painful and much more practical than the majority of the other GN classes. (The quirky names science professors give their classes to entice students to enroll are often cute, but always reek of desperation.) But secondly, I want to become more “science literate”, to borrow Andrew’s term, so I can dig a little deeper into scientific conversation with my girlfriend, who just entered her first year of medical school this fall. I know my technical knowledge of the sciences will likely never be quite up to speed with a science major’s, but learning some off-the-wall fun facts – and I do love my fun facts – could allow me to drop some unexpected knowledge bombs in a conversation. 
As for why I decided not to major in the sciences… Science was always my favorite suject throughout elementary. In particular, Mr. Kase in 6th grade stands out as an excellent educator. We would have a number of mini-labs in class throughout the year where we examined things such as the Doppler Effect or liquid density. Although we probably couldn’t explain much of the technicalities about what we were doing, we were surely discovering the world around us through the lens of science. 
That all changed in middle and high school. My natural curiosity was buried under a pile of numbers and terms for which the practical use was never adequately explained. Fields where my creativity and logic skills were more readily used jumped to my attention. I saw marketing as a potential avenue to a career in the arts industry – even as a man who doesn’t necessarily possess professional-level musical talent. So far, all has been going to plan: I just completed a summer internship with Disney Theatrical Group, on Broadway, where I helped the marketing and planning of “The Lion King” national tour. I can’t say I am disappointed with my major choice so far, but I am hoping to at least gain more of an appreciation for the sciences through this class.
So that’s all for now! Since I need a live link, and I’m on a major Instagram kick lately, feel free to follow me here: http://instagram.com/mjs5842 . I hope all of your semesters are off to a great start, and I can’t wait to meet you all.