Writing Assignment

           I was fascinated after reading the two articles from New York Times about David Brooks talking about how data affect us in our everyday lives.  I learn a lot about what data could and couldn’t do and how important they are.

In one of the article, which is title “What data can’t do”, David Brooks  said that in the statistic world data have a lot information that it can’t prove or explain in this world. For example, he mention about an owner of bank who couldn’t decide whether he should remove is bank from Italy or not.  In the end the owner decided not to remove it. This example was just to prove that we didn’t need data to make such decision. He even gives different situation in which data are not needed to make decision and sometime they give us hard time to deal with. According to David Brooks, we don’t need data to tell stories like writing a novel. Sometime the more data that appears the harder it gets to find what we were really looking for. In some case the bigger a problem occurs the more troublesome  big data also become. Brooks also say that data prefer to focus more on what people like than what the people don’t know much about and that sometime it gives unclear result. This article talks about what data can’t do but in the other article title “The philosophy of data” prove that data are quite important in society.

In the second article, according to David Brooks data helps us explain some things we wrongly understood. For instance, most baseball supporter thinks that when a player makes six or more consecutive shoots, he got the groove. However,  in statistics the data prove that the player  just had luck because the player who missed the six or more  consecutive shoots could also make shoot at the next  round. The same goes for politic election too. People think that the more money you use on your election the more votes you’ll get and that thought was also wrong. It was actually proven that it does not  affect the vote at all. Data could also help us characterized people personality such as being egocentric or not. In one word, data could give us result in quantity and base on that we could make a conclusion.

I was fascinated when David Brooks mention being able to tell when someone is lying or not. Just by listening to the different kind of word how many how many time  the person say it you could tell what kind of person he or she is. It so fascinating because it could help the voter chose their candidate during political election.  For the baseball example, I admit that I also thought that when a player makes a certain numbers of shoot in row had the groove. With this I actually understood that it was only just luck, I’m of kind disappointed because it always made me think that the player was special or had some kind of superpower or something like that.

Describing someone behavior based on data seams to interest me the most because in Brooks article “The Philosophy of Data “, I was amazed that someone personality could be recognized based on data. The further I read his article the curious I became.

I found a lot different interesting subject on this topic. While going through my research I found out that behavior could be distinguish from facial feature base on data. I also found that in some area women are better than men vise versa.

According  to M.Farouk Radwan who wrote the article “Face reading” say that we can judge someone personality from his or her facial feature by using data. For instant, base on data people who look alike are to likely have the same behavior just like babies who almost look alike. While ready the article “Personality traits and university program” written by Beauchamp, Kristy and McKelvie, Stuart J, I  found out that women have  higher score when it come to conscientiousness, industriousness and empathy but men have a higher score when it come to independence. so means that women are sensible than men.

According to Brook, behaviors could be distinguished based on data and from my research I learn that it was possible to do so. In his article it say that its possible tell someone personality by observing the different kind of word he or she says and how the person say it. Before reading the Brook’s articles I thought it was would be difficult to judge someone personality with data but now, it makes sense.

The two articles made think if data were used to determine other thing that we are not aware off.  It made wonder if fortune teller use this strategies to be able to tell or guess someone future. I also wonder if statistics are also use in the game like wheel of fortune, jeopardy, wipeout, and more, or if it also use in the small companies to allow them to make their companies a better place to work at. Now that think about, statistics seem to be everywhere but we just notice it. Maybe will the data reach the level of allowing us to make personnel decision in the future. We never know, in this world everything is possible.

Part 3 Report on research by _OWOEYE_DAVD

DRAFT – PART 3

       Today we live in the Information Age where we understand a great deal about the world around us. Much of this information was determined mathematically by using statistics. When used correctly, statistics tell us any trends in what happened in the past and can be useful in predicting what may happen in the future. But as Brooks says in one of his article, “there are many things big data does poorly.” One of the problems here is correct interpretation of data and its reliability. As Brooks states in his example of use of words “I”, “me”, “mine”, people with confidence use a fewer of those words and vice versa.[] But we can look at it from other perspective. What if “confident” people just know the psychological effect of those words and, in order to persuade someone, avoid using inappropriate one?  Data can’t register one’s intentions, thoughts or moral condition; it can only display visible events and facts.

The article, “Use and Misuse of statistics,” presented by Harvard Business school, talks that before we use any data we should know how much this is reliable. In order to accomplish that, we should be clear about purpose of using the data, what we want to discover. For example, customer satisfaction survey results in the arithmetic mean or average of a group of numbers and equal, let’s say, 3.5 on a scale of 1 to 5.[] But in reality it could be that no one gave to a product a rating of 3.5. Instead, the responses could cluster around a group of very satisfied customers, who scored it a 5, and unsatisfied customers, who gave it a 1. In this case the mean isn’t the most helpful metric for research.

Real life is more complicated than data report, so we don’t have to take cause and effect by granted. If we do such-and-such, then such-and-such will happen. The desire if not requirement that data must be used with every decision creates paralysis.  In present world we are taught to seek perfection, but sometimes we forget that the one thing more important than perfection is simply progress[]. If everything should base on statistics, we would never have great unexpected breakthroughs in the human history.

With statistics, we can’t prove things with 100% certainty.[] For instance, people, recording survey results, may be dishonest or sloppy in those results.  That question has emerged with survey conducted by two criminologists that has raised doubts about the integrity of the New York Police Department’s highly regarded crime tracking program, CompStat. Relying on the anonymous responses of hundreds of retired high-ranking police officials, the survey found that tremendous pressure to reduce crime, year after year, prompted some supervisors and precinct commanders to distort crime statistics[].

The biggest limit to big data is our ability to interpret it. [] Gordon B. Drummond in his work “Data Interpretation: Using Probability” talks about principles of data interpretation. One of his key points is that we should ensure that a sample studied was right chosen and random. As Drummond states, “It’s possible that some scientists are not even clear that the word ‘‘sample’’ has a special meaning in statistics, or understand the importance of taking an unbiased sample,” we can never be certain that a sample will exactly reflect the properties of the entire group of possible candidates available to be studied.  Drummond suggests planning study, establishing hypothesis and estimating the probabilities that the observed data could have occurred by chance, “A properly designed study that aims to answer specific questions will have defined outcomes of interest at the outset, before data collection has started. These questions are then recast as hypotheses that need to be tested.” At the end, when we draw a conclusion, we should consider that absence of evidence in any study is not evidence of absence. If we can’t detect or analyze something, there is no prove it doesn’t exist.

Brooks in his “What Data Can’t Do” publication states that raw data have been structured and analyzed by people who use their own values to draw a conclusion. I can’t disagree with this point. Computers can collect the data but only human beings with their own prejudices, gaps in education and sympathy to a certain things will draw final conclusions. People can’t be 100 % impartial. They will always see things through one’s life experience and can easily bend the accurate data sources by simply asking and changing the question to suit their end goals.

In spite of all flaws, everything we do now in modern world is data driven. Weather forecasts, academic success, politics, stock market, etc. – it all depends on data analysis, which is the best way to understand the present and the past. Big data has its uses, but we should remember that just because we have a lot of data doesn’t mean we have the right data to answer a particular question. Data is only useful if it is honestly and thoroughly gathered. When we are summarizing and interpreting data we shouldn’t blindly rely on raw facts. In order to understand and predict the future outcomes we need to see the problem from different aspects and be maximum objective.

Evans_Dion_Statistics_Writting

Lweb Pt3

Omar R

Brook’s key point is that data can tell more things about situation or specific subject matter than any tool without bias. Moreover, this helps evaluate the situation at hand with a clearer view, with data we can analyze certain pattern that are in the news, on people, nature, politics, economic, education, and much more, the list goes on. However, he also argues that data not always gives a clear cut, as one can say, or a defined answer. Sometimes, you have to trust your instincts and senses. One can argue that with statistical tools to analyze any set of data is going to give a define answer; however, there are moment where the best choice one can make is based on ones’ personal intuition.

Working with data gives a greater tool to know how you are going to perceive the world in this modern age. Everything is computerized and all that information and all that data can give a wider view of your situation. The thing that is interesting is that data can be combined with one’s own instincts and sense of guidance to analyze and evaluate data. There always going to be an experiment where data and statistical results give you one answer but instinct gives you a different answer; with these two (data and instincts) one can have more advantage to come to a conclusion by evaluating all the key points and factors.

Every day, we see how data and statistic is been used to enhance our ability to analyze information and how we interact with it. I am most interested in learning about our capacity on using data and statistical tools to learn new things better than before and how we are susceptible to small changes to cope with new information. We conduct and analyze data from experiments on how the human being learns new things. However, there have not been a lot of implementations to education from these experiments. The only result we can see are the ones where education is provide online (on the Internet), where students and nonstudents take classes online (virtual classroom or online classroom). We have seem a great interest on online education, however, how can we measure the results of the data that is out there? And, is it better to take classes online (sometimes in our own pace) than to be sitting down, listening, and interacting with an instructor face to face in a classroom? I think we can take advantages of both. Right now there top universities taking advantages of this by giving classes online for free to the general public. Then data on how students learn on this online classroom is than evaluated and applied to real classrooms in their campuses. Data is one of the keys to understand ourselves, but we too need to learn on how our instincts are capable to evaluate things on its own and enhance it in the process.

The reason why I am interested in this topic is because human beings need to learn lot of different things throughout life. Since the day we are born, we are bombarded with information that is around us until the day we die. I think the technology we have in our disposal, will help us understand and learn better and faster than ever before.

The kind of information that I found on this topic is about how data mining help students and teachers throughout all level of education achieve better result to better help education to be more effective and improve the system that we have in place, right now.

Through the United States and the world the education system focuses on given standardized tests while students are in the school to measure the output (capability) of the students. This type of data gathering is obsolete because these tests are so narrow (and sometimes so board) that they do not cover all the subjects (or interest that a student, for example, has) and also that the new tools, like virtual classroom, really are better in predicting the achievement of students throughout school. The relationship between this and Brook is that, we may not be listening to the data the way we should to improve education.

In the matter, “schools have changed. In the last decade alone, technology innovations have created new ways of teaching and learning, far away from the convention of textbooks and other printed resources. How can you grasp what your students need now, and really evaluate their progress? Data. Nowadays, one cannot undervalue the importance of data” (Powe, 2011). In the case of Florida’s Broward County School District we see an implementation of data mining system, like IBM AS/400 running IBM’s DB2, to visualize and analyze the student’s performance throughout the day. Teachers and administrators enter information, for example, “historical view of each child’s academic performance” (Messmer, 2000), of the students in the system in a day to day basic. In addition, this demonstrates one way in which data can help students and teachers improve learning and focus on the areas that are the most weak for the students.

Works Cited

Bibliography

Messmer, E. (2000, 02 29). Data mining student performance. Retrieved 04 01, 2013, from CNN.com: http://archives.cnn.com/2000/TECH/computing/02/29/high.marks.idg/hs~index.html

Powe, J. (2011). How schools use data to improve student achievement. Southeast Education Network (SEEN) SEENmagazine (Winter 2011), 39.

Omar Response Piece – part 3

Jonathan Ciabotaru

research

Data Analysis derived from both mechanical and traditional methods help to make better decisions.  An analysis based solely on one of these methods leads to skewed results in the direction of the purpose of the method.  In the traditional method, the emotional or humane factors are captured whereas in the automated method that uses computers and algorithms give a measurement report omitting the emotional or social variables.

In the first article entitled “The Philosophy of Data” Brooks indicates statistical analysis reveals new patterns that humans fail to notice and that it illuminates ideals that humans hold dear which negatively impacts analysis because these ideals do not reflect reality.  Brooks’ basketball example of hot and cold streaks reveals how people’s intuition is wrong when gathering data.  Brooks points out that data gathered from ‘Gilovich, Tversky and Vallone’ shows “that a player who has made six consecutive foul shots has the same chance of making his seventh as if he had missed the previous six” Brooks (1).  In Brooks’ second article entitled “What Data Can’t Do” he gives an example of a Chief Executive of a Bank that had to make a decision about remaining in Italy which had a weak economy at that time.  Brooks’ example shows decisions based on one method of analysis, in this case mechanical analysis, was not enough for the Banking Executive.  Instead the Executive decided to remain in the weak economy “and ride out any potential crisis, even with short-term costs” Brooks(1) because of the emotional and trust connection the Bank had gained from the people of Italy.  Brooks went further to conclude that the analysis is “good at some things and not at others” hence decisions made solely on one method of analysis will be done based on misguided reports because one method only yields to skewed data in the direction of that method.

Applying both the modern computational way of analyzing data along with the traditional methods of human interaction gives a much more fruitful result.  Although the computational analysis leads to promising results based on the quick analysis of large datasets, there is still more work to be done to capture data in better manner.  Stewart et al, research shows the pattern recognition algorithm they used yield results that “investigators can now rely upon alternative sources and techniques to corroborate information about public health events.” Stewart (8).  However, Stewart acknowledges that although the Statistical Pattern Recognition algorithm was good for this purpose it raises more questions about the quality of the core indicator or variables used to determine the pattern.  Stewart et al plans to assess “the quality of an indicator based on such factors requires a more robust qualitative evaluation with input from domain experts” Stewart (8).  In another research about analyzing data over the internet it shows that “non-traditional variations are needed to cope with the unique nature of the internet and its content” Lewis (35) while “algorithmic analysis of content remain limited in their capacity to understand latent meanings or the subtleties of human language” Lewis (35).  Here the content analysis is questioned when large datasets are mechanically analyzed because software has its limitations.  Software is not able to interpret every human instinct especially the emotional behavior as Lewis et al research quoted ‘‘The chief disadvantage is that the computer is simply unable to understand human language in all its richness, complexity, and subtlety” (Lewis (38).

Although progress is made from analyzing data from large datasets the analysis also shows that the variables used for such an analysis can change rapidly and will impact the analysis.  Even the research by Stewart et al agrees that future work needs to be done to ensure the report can offer a better analysis since the qualitative assessment was based on certain variables and in some cases those variables would have had to already been in existence for comparison.  Brooks’ second article also shows that analysis of big data is good and can help with decisions but making decisions in the direction of the analysis will be skewed decisions since such an automated analysis does not cater for the values human have for another.  Indeed Lewis research supports hat computers has its limitations and analysis based solely on this will be skewed because of these limitations.  Decisions based on analysis done by both automated and traditional methods give a clearer picture to the decision maker.

Works Cited 

Brooks, David. “The Philosophy of Data”. New York Times 4 Feb. 2013: A23. Web. 1 Mar. 2013. <http://www.nytimes.com/2013/02/05/opinion/brooks-the-philosophy-of-data.html>

Brooks, David. “What Data Can’t Do”. New York Times 18 Feb. 2013: A23. Web. 1 Mar. 2013. < http://www.nytimes.com/2013/02/19/opinion/brooks-what-data-cant-do.html >

Lewis, Zamith and Hermida. “Content Analysis and Big Data”.  Journal of Broadcasting & Electronic Media Mar. 2013: p34-p52. Web. 15 Mar 2013.

Stewart, Fisichella, Denecke. “Detecting Public Health Indicators from the Web for Epidemic Intelligence”. 2010. Web. 15 Mar 2013. <http://www.l3s.de/web/upload/documents/1/paper44_Stewart.pdf>