-
Mark Sabat wrote a new post on the site MIS 0855: Data Science Spring 2017 7 years, 4 months ago
Leave your response as a comment on this post by the beginning of class on March 23, 2017. Remember, it only needs to be three or four sentences. For these weekly questions, I’m mainly interested in your o […]
-
Mark Sabat wrote a new post on the site MIS 0855: Data Science Spring 2017 7 years, 4 months ago
Here is the exercise.
And here is the spreadsheet to complete the exercise [In-Class Exercise 8.2 – OnTime Airline Stats [Jan 2014].xlsx].
-
Mark Sabat wrote a new post on the site MIS 0855: Data Science Spring 2017 7 years, 4 months ago
Some quick instructions:
You must complete the quiz by the start of class on March 21, 2017.
When you click on the link, you may see a Google sign in screen. Use your AccessNet ID and password to sign […] -
Mark Sabat wrote a new post on the site MIS 0855: Data Science Spring 2017 7 years, 4 months ago
Here is the exercise.
-
Mark Sabat wrote a new post on the site MIS 0855: Data Science Spring 2017 7 years, 4 months ago
Leave your response as a comment on this post by the beginning of class on March 9, 2017. Remember, it only needs to be three or four sentences. For these weekly questions, I’m mainly interested in your o […]
-
I have made one of the Excel mistakes it was to Miss the data type. I was working on analyzing data to see in what months there was the highest customer sales for a company, but the data type was considers a data so the analysis was inconclusive. I had to use my back up copy of the data to continue the analysis because it messed up the data set.
-
I have never made one of these Excel mistakes, but I think the most important to avoid is Number 4, which is sort a spreadsheet but not include all the columns. I think this is the most important because when you select all they only select certain columns and some of the data won’t be included at all. The most important part of your work is the data.
-
Typically I don’t have many problems when using Excel, but one mistake on the list I have made is #9: Copying formulas that use relative coordinates. This seems to be a common mistake among college students. One time when I created a grade book for a course, I forgot to add a “$” before the cell location which corrupted my intended calculations.
-
I don’t use excel very often, but I have made mistake Number 7: Put Values in Field that are Supposed to be Pointers or References. I made this mistake when I was working on a budgeting project. When doing the calculations to determine my expenses, I put in criteria that helped me get to the sums, but I put this data under the column title “Final Expenses.” This was an issue because it effected my overall expenses greatly, and I had to go back and fix my Budgeting Project.
-
I have never made one of these mistakes but I think it is important to copy formulas that use relative coordinates. If you use the wrong coordinates in your formulas your answers that you get out of the formula will be wrong and mislead your work. The ability to navigate excel is so important to be able to do any type of work inside of excel.
-
I do not work with excel much using databases, I usually use it to make a form or a quick spreadsheet. I think one of the most important mistakes not to make is number 3, start working on your data without doing a full back up first. I can recall even in in class exercises students were having trouble with excel and it freezes, technology can not always be trusted. It is important to back up your work with anything.
-
As a Computer Programmer and a student I have done programming with lot of languages. The most common mistake I do is with using the wrong syntax and data type. It is very important for a programmer to think about your variables and instances you create. Also with syntax, same with tableau I have used wrong brackets, and also for instance print statement might have been missing semi colons or quotation marks. Without the debugger it is very hard to find those error and correct those.
-
I have never made any of these mistakes, but I think the data type is very important. Changing dates to integers could really be confusing to people and can mess up your data a lot. I also things not being careful of the columns you use is very important and could made a big mistake in your data as well.
-
I am guilty of committing a few of these excel issues but certainly the issue I have had a problem with most often is missing the data type and mixing up integers with dates/times. I would argue that this is one of the most important issues to avoid as well, because it could end up skewing the entire data set. The other issue that I frequently commit is clicking “yes” without carefully evaluating the message that says “do you want to remove this from the server?”. I think this becomes a problem not only because the dialog boxes are not specific enough, but we tend to just click through the prompts to get to the next step quickly, though sometimes at the cost of efficiency.
-
I have never made any of these mistakes, but the most important one from all of these is: changing dates to integers can be confusing to people and can mess up there data and excel sheet a lot.
-
I have absolutely used a deduping tool with “loose” criteria first. It was terrible and I’m lucky I caught it. I was importing data into our Salesforce tool at work and I didn’t realise that I had almost removed 1/5 of the data because my criteria did not include first AND last name, just last. As you can imagine in a list of 5k clients there are a fair few Smiths, Allens, etc.
-
Yes I have made a mistake and I made the most common one you could make. When I was working on a basic excel assignment in high school I had finished and when i went to clean it up, i clicked yes and all of my data got corrupted.
-
I personally have not made any of this mistakes before but the one I think that is most important is #4, the data type. If your dates change to integers that can screw up the rest of the data and make everything hard to comprehend.
-
I feel like if I used excel on a more day to day basis, I could definitely be prone to making one or some of these mistakes but the lack of Excel use makes it hard to say I have made any of these errors before. The error which would be most important in my eyes is probably case 4 that the author presents which states “Number 4: Sort a spreadsheet, but not include all the columns”, the reason being, without an entire column the data analysis will be highly skewed and not an accurate representation of what one is trying to find out.
-
I haven’t had to deal with any of these problems yet but if I had to say which I thought was the most important one to be wary of, it would have to be Problem #10.
“Number 10: Open a CSV file directly into Excel”
If your data is corrupted from the start, you have to take steps to fix it, or if you anticipate compatibility problems, prevent it otherwise all the work you did would be for naught. -
I haven’t made a mistake listed within the article, but I believe that #3 is the most important thing to be conscious of when working with Excel. The backup is essential to either refer back to revert back to. Without the backup there is no safety net.
-
The mistake I made was the miss the data type. I do this often since date and times are converted to long integers and make a confusing mess on the spreadsheet. I believe this is an important one not to mess up because it can be a hassle to fix the problem.
-
I have never had a problem using excel. The important mistake people should avoid is number 6. An error in data type can cause a lot of problems when analyzing data.
-
I have made a couple of the mistakes listed in the article. Number 9, I’ve copied formulas that use relative coordinates for my MSOM class and it threw my answer off but i was able to find the mistake and correct it. I also made a mistake of number 3 working database without doing a full backup first.
-
The mistake that I have made is #6: missing the data type. As I entered the dates into Excel, it keeps understand my dates as integers, which creates some problems for my Excel spreadsheet. Another mistake that I have made is not doing the back-up before using the Excel file, which I think is one of the most important ones to avoid.
-
I have not made one of those mistakes, but I think the most important to avoid for me would be number 9. For my job I do a lot of data entry into excel for clinical research studies, and formulas are almost always needed to interpret the integer data types. And often formulas are embedded in the spreadsheet itself. Formula errors will lead to misleading research results and could be catastrophic to the reputation of my department and the institution (Temple Hospital) that publishes it.
-
Without question the biggest mistake I make when using data is that I click yes without reading the message attached about removing from the server. I too hastily skim through messages and try to speed through my work which actually causes me too slow down. I think this is a very popular problem, people in the world today are in a rush to work and sometimes don’t read everything they need to read.
-
I have not often used Excel, but the one I’ll be most sure to avoid is number 4: sorting a spreadsheet without including all of the columns. From what I have learned thus far about Excel, being really sure of what you have selected is crucial to not making any mistakes. After sorting a spreadsheet and later finding your calculations are incorrect because of not having selected all of the columns could be hard to spot errors and cause me to have to redo the entire sorting all over again.
-
I don’t use excel that much. In my opinion the mistake number 6 is the one that could happen the most.This could lead to a serious mess in your data confusing and mixing it in a way you don’t want it to.
-
Missing the data type, in my opinion, is the easiest mistake to make and the most harmful, since it a erroneously computed value can skew an entire data set.
-
I’ve made the mistake from “number 6: Miss the data type” quite a few times in Excel. I’ve had number of data entries accidentally be entered as dates instead of integers. Specifically, I’ve tried to enter a value and it automatically switches to a date and then changes the entire function outcome without me realizing because, well, I knew I entered the data correctly and did not expect Excel to suddenly switch what I typed as soon as I clicked Enter to move on to the next data entry space.
-
I have never made any mistakes, but I think the data type is important. I think that being careful of the columns you use is very important and could made a big mistake in your data if you are not careful.
-
-
Mark Sabat wrote a new post on the site MIS 0855: Data Science Spring 2017 7 years, 4 months ago
Here is the exercise.
And here is the dataset you’ll need [Vandelay Orders by Zipcode.xlsx].
-
Mark Sabat wrote a new post on the site MIS 0855: Data Science Spring 2017 7 years, 4 months ago
Here are the instructions (in Word) (and as a PDF). Make sure you read them carefully! This is an assignment that should be done individually.
And here is the data file you’ll need: VandelayOrders(Jan).xlsx.
-
Mark Sabat wrote a new post on the site MIS 0855: Data Science Spring 2017 7 years, 4 months ago
Some quick instructions:
You must complete the quiz by the start of class on March 7, 2017.
When you click on the link, you may see a Google sign in screen. Use your AccessNet ID and password to sign […] -
Mark Sabat wrote a new post on the site MIS 0855: Data Science Spring 2017 7 years, 4 months ago
Here is the exercise.
-
Mark Sabat wrote a new post on the site MIS 0855: Data Science Spring 2017 7 years, 4 months ago
Leave your response as a comment on this post by the beginning of class on March 2, 2017. Remember, it only needs to be three or four sentences. For these weekly questions, I’m mainly interested in your o […]
-
http://www.nbcnews.com/id/43499438/ns/business-consumer_news/t/whats-all-these-data-breaches/#.WLM8_xiZMdU
This article is based on the security of our personal information. They talk about how terrifying the Citibank and Sony hacks were and emphasize how knowledgeable hackers are becoming. This article questions the actual security of our personal information and if it is really secure or not. -
https://fivethirtyeight.com/features/oscars-night-was-predictable-until-the-very-end/
This article talks about this years top picks to win the Oscars, compared to the pictures that actually won. They used a model that assigned points to nominees as they win awards that have historically been predictive of Oscar wins. This year there was a big upset for best picture, which the model did not correctly predict. It is interesting to see how we can use predictive models based on past data, especially in current events such as the Oscar’s which is one of the largest broadcasts of the year.
-
https://www.scientificamerican.com/article/will-democracy-survive-big-data-and-artificial-intelligence/
This article talk about how the the evolution of technology with a shift in big data and artificial intelligence may effect how society is organized. A major concern that there will be a trend that goes from programming computers to programing people. It talks about the principles that help prevent this concern like decentralizing information systems, improve transparency to increase trust, support collective intelligence, and more. -
https://climate.nasa.gov/evidence/
This article talks about the changes in climate over the past thousand years. It provides data that shows our current carbon dioxide level compared to years past and how high it is in comparison. It also provides data that shows how there is less snow coverage on Earth and how the temperature of the Ocean is getting significantly warmer over the past couple thousand years. This data allows us to track how much global warming is actually taking affect. I found this to be interesting because it has been oddly warm the past week for it being February and was curious why this was so. -
In this article, writer Neil Irwin covers economic statistics. I used this article because I am currently in Macroeconomics and I am majoring in Statistics. This article applies to Data Science because it uses data sets to determine the current state of our economy. Irwin states that the quick rise and slow fall of unemployment rates in the past decade have finally leveled off to where they typically sit and sat in early 2007, prior to the boom and bust catastrophe that shook our economy during 2008, 2009 and 2010. Unemployment rose as high as 10% during this recession, nearly twice the current rate.
-
This article discusses how a new type of malware called “ransomware” is beginning to become more popular among programmers and is now also targeting previously unaffected Apple users. I find this fascinating, not only because it targets Mac users which are gaining in market share, but that it also holds your files captive – hence the term “ransomware”. It encrypts your files and the malware creator demands untraceable bitcoin payment to unlock your files. The software by design does not create an encryption “key” so there is no way to gain back your information. This is not only directly targeting Mac users, but more specifically people who torrent media.
-
I found the article “Apple vs. Google company structure, as seen through patents” to be interesting because they are known to be some of the most innovative companies. In this visualization it compares patent ownership which provides a good example on the companies structure. It shows that “Over the past 10 years Apple has produced 10,975 patents with a team of 5,232 inventors, and Google has produced 12,386 with a team of 8,888”. This can be related to the more controlled system at Apple compared to the independent system at Google. -
https://www.wsj.com/articles/youtube-tops-1-billion-hours-of-video-a-day-on-pace-to-eclipse-tv-1488220851
This article talks about the growth of YouTube. The article states that YouTube just surpassed 1 billion hours off video on their website, with all of those hours of data comes large amounts of data. YouTube deals with such larges amounts of data and they store all of it which can seem rather mind blowing. The storage of all that data is very expensive and that is why it is unclear if YouTube is actually making a profit. The article also has a info graphic showing the growth of internet viewing is growing versus the decline of television. -
https://fivethirtyeight.com/features/kemba-walker-doesnt-care-how-close-youre-guarding-him/
This article from FiveThirtyEight uses advanced statistics to detail the play of Charlotte Hornets point guard Kemba Walker and how his three point shot has improved over the years. Being an avid NBA fan, this article was very interesting to me, as it looked at his play in a different light than normal. The two statistics it used was 3-point percentage and percentage of time his defender went under a screen. To elaborate, when that second figure is lower, it basically means that his defender is covering him closer than normal. What these metrics showed was that over the years his defenders have covered him tighter but his long distance shot has improved. Very interesting in my opinion. -
https://fivethirtyeight.com/features/trumpbeat-is-trump-already-messing-with-government-data/
In this article, the author questions how Trump’s administration is planning on measuring net exports. It is interesting because it ties right into class and because I love economics. The article includes a lot of data. For example it tells the reader that since 2010, exports are up by thirty percent, which is a lot less than what Obama promised it would be. It then shows how the data one chooses can give biased results. For example, Trump’s administration counting “re-exports” as imports but not exports would make our deficit look way bigger than it is now. Then, any positive change in the deficit would seem huge. It is quite obvious that the reason Trump’s administration wants to change the way that they calculate net exports because they want to make it look better. -
This article is about the introduction of sensors into sports for the use of collecting data. Sports like football and baseball are currently the leaders in terms of interests, they would use these sensors to measure strength of impact in football, and bat speed, pitch, and trajectory of a hit in baseball.
-
http://fivethirtyeight.com/features/whos-trying-to-buy-an-academy-award/
The articles lets us estimate how much financial support various productions are getting when it comes to wooing the Oscar.
It includes a data table showing the biggest print campaigners, ranked by total square inches of advertising bought in the two trades. -
https://www.wsj.com/articles/SB10001424052702304692804577285821129341442
This article talks about data and how insurers use data to analysis how data might predict future outcome with high cost patients. Their has been a growing number of insurers adopting this data related technique and this article describe this accurately.
-
https://projects.fivethirtyeight.com/soccer-predictions/?ex_cid=rrpromo
This is a data set that gives you the predictions for the top European soccer leagues going on. It gives the percentage for each team to win the league, the percentage they have to win their games coming up, and the percentage for the team to qualify for different European tournaments for next season. I check this database at least three times a week to see how my favorite teams stack up in their league and their up coming games.
-
This story uses data visualizations as well as massive amounts of data from the stock market to calculate the projection of the stock market. The stock market is booming — reaching all-time highs– and it’s really interesting to see how reporters can calculate among all of the different stocks, the direction of the market.
-
In the article “Mass Spying Isn’t Just Intrusive—It’s Ineffective”, Granick points out that even tough American agencies have been performing massive spying for years, that hasn’t stopped terrorist cases from happening. She mentions the Boston marathon bombing, the Charlie Hebdo massacre, and a failed attempt to detonate a bomb in an airplane as examples of cases where the government had previous information that the people who performed these terrorist acts were dangerous, but failed to further investigate. -
https://fivethirtyeight.com/features/the-earth-in-a-suitcase/
I found this article very interesting in the fact that how complicated it is to leave the earth. To explain, most people do not think how much thought goes into making life easier for astronauts. To be fair, going to the moon is different than going to Mars. This article mentions all the things travelers would need when the get to Mars.The most interesting part of the checklist is the need for psychological help. The authors collected qualitative data about the mental status of an individual outer space, away from human contact.
-
http://www.wsj.com/livecoverage/snap-ipo
This article provides us with lots of interesting data about Snaps initial public offering. It compares Snap Chat’s opening price of $24 per share to other hot tech IPOs in previous years, such as Facebook and Twitter. It also provides lots of valuable data and charts comparing Snap’s IPO to other historically popular IPOs. -
https://projects.fivethirtyeight.com/trump-approval-ratings/
This chart is interesting because it shows the rates of approval that trump has.It is interesting because it shows different sources, dates and also sample sizes.On average the dissaproval to trump is bigger than it’s approval.An interesting fact is that there are some sources that put Trump approval higher than it’s dissaproval. -
https://fivethirtyeight.com/features/people-are-faxing-their-senators-up-to-300-times-an-hour/
This article provides data about faxes from only one online faxing service, yet still provides substantial numbers. It details the amount of faxes received in between January 4th to February 13th showing that on average, our beloved Senator Pat Toomey, had received around 300 faxes an hour. The senator has received over 20,000 faxes, two times as many as the Senate Majority leader Mitch McConnel and six times more than his fellow members of Congress. Toomey’s office says that they treat and read every single of these faxes, but is there any truth in that? And if they are read, do they contain any influence? -
This article talks about how the finance world is leaning more and more on big data. They use this data to determine if you are creditworthy or not by your tweets, facial expressions or Facebook status. The big data can also determine if a person is going to die early or if they will die from some other causes. This allows insurance agents to price the premiums off the data that they retrieve. Big data is frequently used in the retailing industry as well to determine consumer trends and preferences.
-
https://fivethirtyeight.com/features/firing-claudio-ranieri-wont-fix-leicester-city/
What I found interesting about this article is how data was used to analyze the success of a football club. I agreed with the author of the article in that, I do not believe it was the correct decision to sack manager Claudio Ranieri. The data showed that managerial changes don’t directly affect the teams performance too often. Leicester had a 5000-1 chance of winning the title and went on to beat the odds under his reign.
-
-
Mark Sabat wrote a new post on the site MIS 0855: Data Science Spring 2017 7 years, 4 months ago
Some quick instructions:
You must complete the quiz by the start of class on February 28, 2017.
When you click on the link, you may see a Google sign in screen. Use your AccessNet ID and password to […] -
Mark Sabat wrote a new post on the site MIS 0855: Data Science Spring 2017 7 years, 5 months ago
Here is the study guide for the first midterm exam.
-
Mark Sabat wrote a new post on the site MIS 0855: Data Science Spring 2017 7 years, 5 months ago
Here is the exercise.
And here is the graphic file you’ll need: Philadelphia Area Obesity Rates.png.
Right-click on the file and save it to your computer.
-
Mark Sabat wrote a new post on the site MIS 0855: Data Science Spring 2017 7 years, 5 months ago
Here is the exercise.
Before you start, save this Tableau file and the studentloans2013 Excel workbook to your computer. Remember, to save the file right-click on the link and choose “Save As…” (don’ […]
-
Mark Sabat wrote a new post on the site MIS 0855: Data Science Spring 2017 7 years, 5 months ago
Leave your response as a comment on this post by the beginning of class on February 16, 2017. Remember, it only needs to be three or four sentences. For these weekly questions, I’m mainly interested in your o […]
-
Although all eight principles are important in their own way, I think being skeptical of the data is crucial. As Few notes, too often we fail to question the information that we get from data because it is easier to just take things at face value. If we are not skeptical of what the data tells us, it isn’t as beneficial as it could be and we aren’t learning all that we can from it. Data offers us explanations but without questioning it further we cannot find solutions.
-
All of the eight principles are important and helpful but I think one of the most important one is simplify. This is because no one wants to look at a data visualization which is complicated and difficult to understand. The main reason that people visualize the data is to make it easier to comprehend not to make it more complex and confused. So the more simple and elegant of the data visualization, the better results it will come.
-
I think all of the eight principles are important when they work collectively together. Looking at just one of the steps I believe that view diversely is extremely crucial to data analysis. From data that is collected you can draw different conclusions by looking at the data in different ways. It also enables your to link insights together to form a puzzle that will help to answer your proposed questions.
-
I believe that all 8 principals are important to data visualizations. However out of those eight principles I really feel that the principal of explore is the most important. Explore makes us use data and explore within it rather than just look for a specific answer. The purpose of data is to find out new things and explore .
-
I believe that each of the 8 principles is important but I think the most important would be the one about simplification. Simplifying data means that someone must be able to understand the data and “dumb it down” enough to make a visual that anyone would be able to understand. If the data is not simplified, the visualization will not be successful.
-
The principle I find the most important is to be skeptical. I know that I am the type of person to read or look at something and simply just believe it without any questions asked, but who is to say that what I am seeing/hearing is true? I feel that being skeptical and trying to understand where the data and information is coming from is a necessity for data visualization.
-
The principle that I find to be most important is simplification. When dealing with complex, messy data, it is nearly impossible to come away with a firm conclusion which is why I find this principle to most important. For instance, if you were given a set of data that included statistics of every NFL team, it would be difficult to compare and contrast where your favorite team stood. However if one were to simplify the data by creating a bar chart of total points scored, it would be very simple to come away with a conclusion.
-
All of the 8 core principles of data visualization are equally important in my opinion because they work hand in hand on clarifying good data. If i had to pick the most important/valuable principle I would say be skeptical. The saying “There’s two sides to every story” goes along way and I think this relates to being skeptical. It happens too often that the first thing we see we believe to be true or useful, but thats not always good. Some information that is hidden could be more useful than the information right in front of us and that is why we have to be skeptical and research deeper into certain data to get the most useful information.
-
In my opinion the most important of the 8 core data principles is to “Attend.” I found the gorilla and ball video to be a very interesting example, and the analysis in the linked article was also valuable. It is incredibly easy for us to focus on unimportant data points especially when we are going in already looking for a trend we may miss something glaringly obvious.
-
I would say the most important principle is the compare principle. This is because you have to be able to be able to compare your data side by side. If you depend on your memory then your not going to be remember every number an effectively go through them.
-
In my opinion the most important principle of data visualization is to compare. If you have a set of data and nothing to compare it to, there is no point in having it. Comparing numbers and data is what we do in order to understand data, make future assumptions and be more efficient as a company in future decisions.
-
I think that Simplify is the most important of the 8 principles. This step basically challenges the artist or data designer to give the audience the most important or relevant data via an easy delivery. Certain graphics are going to require multiple categories of information and a lot of extra or unnecessary data. By simplifying the data, the audience can quickly sort through information that is relevant to them. This step increases efficiency.
-
The most important core principle, in my opinion, is “Simplify”. I believe that data visualization is storytelling. The author of the visual is trying to tell the reader a story, or some core meaning. It is hard to achieve that by complex visuals, legends, colors, etc. However, at same time, i think one should not fall into the trap of oversimplification. There is a balance that is needed. Over simplified is too light on data, while properly simplified data is the most efficient.
-
Out of the eight core principles, I believe, “ask why” is the most important one. It is easy to know what’s happening but to get more detail and become more knowledgeable you need to ask why it’s happening. This will help you develop better knowledge about the specific data and help you become more interested. Knowing what’s happening is the easy part but to actually become an expert, you need to ask why it’s happening.
-
All of the 8 principles are very important when it comes to data visualization but in my opinion the most important principle is “Simplify”. Visualization of data at times can be hard to understand so simplifying the data will make it easier for people to understand and it also increases just how precise the information is because the data is used more efficiently. All of the points tie together when it comes to data but I feel as if simplify definitely stood out in my opinion.
-
Of Few’s 8 core principles for effective data visualization tools, I believe the most important one is “explore.” We are often limited by the specific question we are trying to answer or the hypothesis we are trying to prove. However, it is important for a good data visualization tool to let you “play around” with the data, allowing analysts to discover new insights they were not necessarily looking for. Being able to freely explore the data allows us to take a less narrow approach to data analysis.
-
I think the most important principle is “Explore”. When creating a data visualization, answering one question shouldn’t be your only mission. You should start answering more questions and inform everyone related facts and information. That is what will make your data more useful. It will also let anyone who sees it fully understand the context you are trying to give.
-
I believe the most important core principles Stephen Few notes is “simplify.” So often, I will be reading a story from a national publication and look to the data visualization or infographic to make the connection the journalist is describing more clear. But oftentimes, I read the graphic that has so much extra information that I end up with more questions than I had before. Simplifying to show the most important information in the most straightforward and understandable way is crucial for having an effective data visualization. Without having a simplified graphic that is clear to a reader, then the rest of the core principles will be hard to achieve.
-
In my opinion, the most important principle of data visualization is skepticism. The number of data and analyses out there is infinite. Because of this it would be very closed minded to believe every data visualization one sees. It is very easy to look at something and have an immediate impression on it. However, oftentimes things are portrayed in a way that cause us to be visibly inclined to look at it and believe what isn’t necessarily true or as important as we perceive it. Having a skeptical mindset when looking at a data visualization simply ensures that what we are viewing is worthy of our attention for all the right reasons.
-
In Hoven’s article, eight core principles were: simplify, compare, attend, explore, view diversely, ask why, be skeptical and respond. In my opinion the most crucial of these principles is ask why. Asking why is an important step in any data analysis; just like in this class, before we began looking at data we had to come up with hypothesizes. Asking the wrong question or coming up with the wrong hypothesis can sometimes lead to misuse of data and not proving the right answer. This is why I believe being skeptical is the second most important principle following asking why.
-
I think Hoven’s most important data visualization rule is simplification. Especially when it comes to data visualizations for journalism and in the news, it’s necessary that the audience is not confused by ambiguous information that’s irrelevant to them. The audience should only be provided with the information they need, or else it is simply a distraction.
-
I think the most important rule is simplification.For the person who read the data it has to be easier for them to obtain the data they want without missing anything important. And that job for the person who post the data is clue.Also if it’s simplified it means it will be better to digest and it could attract.
-
All of the rules are valid and important, but I believe the most important rule is to be skeptical. Without proper analysis of the data, you are tempted to arrive at the easiest possible conclusion, which may be erroneous. If you graph the correlation between ice cream consumption in the United States compared with the increase in gun violent you may be inclined to believe that ice cream causes gun violence – they both spike in the summer months. In actuality, people eat ice cream in the summer months because its hot and gun violence increases in the summer months because people are more likely to be outside and interacting at gatherings. If you are not skeptical of the data and the correlation vs causation, you are more likely to arrive at the wrong conclusions.
-
In my opinion I believe “Be Skeptical” is the most important principle. The article goes on to tell that all too often people take the first answer given as gospel and my beliefs align with this theory. With things so readily available to us, people hardly ever challenge or look further into matters.
-
Although all the eight core principles Stephen Few recognizes seem to hold equal importance, personally I think in data visualization, the principles of “simplify” is most important. This seemed most crucial to me in displaying data because if the data is represented in a confusing or messy manner, the message of its importance will be hard to comprehend. Data visualization after all, is about representation and enabling easier interpretation of data collected. Therefore, I think the most important core principle is to “simplify” so that everyone can learn from the visual.
-
I believe the most important one is simplify I believe this because data need to be easy for the consumer and person alanayzimg the data to read it in a matter that it easy and simple. If data is umstructured then it going be hard for us to take the data and make predictions on it
-
The 8 Core principles on data visualization are Simplify, Attend, Compare, Explore, View Diversely, Ask Why, Be Skeptical, and Respond. I think the most important one is to Ask Why? There is so much data in the world today and not all data is accurate to it is important that you are getting a good understanding of the visualization.
-
I think that ‘ask why’ is the most important principle in data visualization. It’s easy to just say that something is happening, but a deeper analysis is vital in trying to make a point across. Why something is happening shows what changes can be made in order to expand on the trend or prevent it from happening in the future, and for real change to occur.
-
In the Hoven’s article, there were eight core principles were: simplify, compare, attend, explore, view diversely, ask why, be skeptical and respond. The most important core principle in the Hoven’s article is “Simplify”. I believe that the meaning data visualization is storytelling, the author of the data visualization is trying to tell the reader a meaning of the story. It is hard to realize that by composite visuals, legends, and colors. Though, at the same time, I think that one should not fall into the deception of simplification. Oversimplified is very light on data, while appropriately simplified data is the most efficient data that we can view.
-
Of these eight core principles, I believe that the most important is to be skeptical. To be skeptical is to be human; while computer programs can accept the data that is presented to them, even if they are built to notice flaws, we can’t replicate the complexity of the human brain in analyzing data.
-
After reading the article, I feel that “simplify” is the most relevant. Data that cannot be processed is useless and I feel that the larger the audience is that data is able to impact, the more effective the data is. The simpler the data is, the larger the audience is that can process it is so therefore the usefulness is increased as simplicity is increased.
-
I think the most important rule is simplification. The data that reading by people has to be easy for them to get the data they want without missing any important point. Also if it’s simplified it means it is easier to digest and it can attract.
-
-
Mark Sabat wrote a new post on the site MIS 0855: Data Science Spring 2017 7 years, 5 months ago
Here is the exercise.
And here is the spreadsheet you’ll need to complete the exercise [In-Class Exercise 4.2 – FoodAtlas.xlsx].
Make sure you right-click on the Excel file link and select “Sa […]
-
Mark Sabat wrote a new post on the site MIS 0855: Data Science Spring 2017 7 years, 5 months ago
Some quick instructions:
You must complete the quiz by the start of class on February 14, 2017.
When you click on the link, you may see a Google sign in screen. Use your AccessNet ID and password to […] -
Mark Sabat wrote a new post on the site MIS 0855: Data Science Spring 2017 7 years, 5 months ago
Here is the exercise
Here are the links in case you cannot click from the document.
History, Economics and Social Issues
Science and Health
English, Fine Arts and Entertainment
Remember to […]
-
Mark Sabat wrote a new post on the site MIS 0855: Data Science Spring 2017 7 years, 5 months ago
Here is the assignment.
Here is the worksheet as a Word document to make it easy to fill in and submit (along with your Tableau file).
And here is the data file you will need to complete the assignment […]
-
Mark Sabat wrote a new post on the site MIS 0855: Data Science Spring 2017 7 years, 5 months ago
Leave your response as a comment on this post by the beginning of class on February 9, 2017. Remember, it only needs to be three or four sentences. For these weekly questions, I’m mainly interested in your o […]
-
On of the most important takeaways I learned from Brian’s talk was that it is important to develop and innovate, but its also very important to be able to market it and convince others it’s a useful tool. I was amazed at the price in which twitter bought his startup and while it can be attributed to the development and innovation itself I was equally impressed with their ability to convince a giant like twitter of the usefulness of his startup. In retrospect mobile re-targeting seems like an easy sell now because everyone does it. However, at the time it was significantly less popular and convincing twitter to make a hundred million dollar risk is very impressive.
-
Brian surprised me with how much time and preparation he put in to launching his start up with Twitter and it actually interested me I was talking to someone who knew so much about mobile targeting especially being that he worked with a social media website that I visit a lot. Brian helped me understand why data is important to sell in advertisements because it gives readers and costumers a look into what they are signing up for and how the “product” is used.
-
The presentation Brian gave us was very interesting. I had never heard of the real time auctioning and how it works is pretty amazing. Brian’s talk was very in depth which it had to be because it was the same presentation he used to sell his company to twitter. I had always wondered how something i just viewed on a different site was now an add on a different website and now i know that its not just based of cookies. The presentation showed us how to really become successful with a startup.
-
I think the most important takeaway from Brian’s talk is that there is a way to work and understand big data. In addition there is a way to capitalize on consumer data. I learned that by using customer data you are able to specialize advertisements towards individuals. Through mobile advertising companies can make a profit and meet the needs of consumers.
-
One of the takeaway that I can learn from Brian Long’s talk on Thursday is that big data can be be used to display ads, which relates to the user’s preference. Big data is especially useful for marketers to promote their products through ads and it also helps the ads itself works well. I can now address my own question why I search some products on Google but the related ads about the product keep popping up on my facebook. In addition, I pretty like the information he gives about mobile deep linking. when user clicks the ads on app, instead of leading user right to the app, an URL will appear and direct user to a specific location within an app -> drives user engagement
-
Brian Long’s talk on Thursday helped me realize the relevance of innovation and understand that big data is used in various manners. I thought it was interesting how he pointed out the various ways to target consumers via big data.It was pretty incredible to see how much twitter bought his start up company for. Overall, I think I learned that big data was used more than I thought for marketing and thus helpful to big mobile companies.
-
One takeaway is that big data is used to compile information to target ads to specific users. The people collecting the data do not necessarily care about the person’s personal information, but their habits online (i.e. – what they like to look at, buy, sites they visit often, etc.). The main goal is to retarget, or get people to view a site or product as a result of an ad they’ve seen. Ad companies bid on open slots on websites to get their ads displayed and hopefully attract more consumers.
-
One takeaway that i got from Brian is that big data is everywhere. In addition, companies used big data to predict and evaluate user preference. The way marketer target crowds is from big data using it allow them to predict success rate for mobile application. He allowed us to understand how a startup is able to connect with it user and make everyday decisions.
-
The most important takeaway from Brian’s talk was how big data can be personalized. I thought it was really interesting that companies can track what you are looking at and then send you ads to other similar things. I can definitely see how that can turn profit because whenever I look to buy something on amazon ads for similar things pop up on different websites and I always end up looking at them and am tempted to buy what they are advertising. It amazes me how big data allows for this.
-
Brian’s talk was a fun break from the normal class structure, he was entertaining as well as very smart and did a great job educating us on big data and how companies use it to their benefit. I though the most interesting part was that with most companies we are kept as a serial number not as a name, this shows that companies aren’t focused on us as humans but are focused on our consumer history and buying tendencies to be more effective. I also found Brian’s story about working with different companies acquiring data fascinating and very informative.
-
Personally I think Brian’s talk was eye opening in regard to the application of big data. The advertisement re-targeting is a very useful tool for both companies as well as consumers. That being said, it still is a bit creepy. However, I do see more value in the use of big data.
-
It was very interesting to hear Brian’s story. Apart from the fact that his company was bought by twitter by a mammoth $100 million, the idea of redirecting ads is something I have not heard of before. Interestingly, when I was on Facebook today, I saw this question that FB asked me. It was asking whether I recalled seeing an Ad from this certain company in the last 2 days. The moment I saw that, the first thing that came to my mind were the topics we discussed in class about big data. More specifically, how big data can be used to target individual more surgically. Facebook probably asked me that question to analyze how well their Ads are targeting users.
-
I would like to be one of those people who can say they loathe target marketing, but I just bought a $40 computer case this morning because of it and I don’t regret it one bit. I agree that it is invasive and it can feel like an invasion of privacy, but it works. Bryan was able to sell his company for as much as he did because dingbats like me with impulsive buying habits can’t say no to the one attractive Facebook ad about a computer case that leads down the rabbit hole of an eventual purchase. I was singled out based on my search history, buying patterns and user input…and I hate that I liked it.
-
The most important takeaway from Brian Long’s talk was how Big Data can be used within professions. It was very helpful that he applied our studies of Big Data and connected them to money. This gave me a great deal of perspective on how important Big Data truly is. I learned that Big Data can be used to sift through results (bids) rapidly in order to find the best option. His mobile advertising company was a massive success, and this was very inspiring and helpful.
-
I felt like the most important takeaway is how our data is being used to target ads to us. It was a very interesting lecture because I have always wondered how these ads would seem to track me. I’m not comfortable with the idea of a person knowing my purchase habits especially if I start getting recommendations based on one thing I only clicked on by accident. That was a headache. What also blew my mind was that there would be an auction of sorts where the highest bidder would be given the ad space and all of it takes place in an instant.
-
The one takeaway that I got out of Brain talking was that big data is the future of marketing. He talked about how companies use the big data that he collected and marketed to a specific demographic because of a concert, game, venue, or any other activity that would bring in marketers. This kinda of big data has been around for that past decade but only now are people starting to utilize it as a special tool for marketing and overall business needs.
-
There were several takeaways I learned from Brian Long’s presentation. I was amazed with the price that Twitter decided to buy out his company for. I was also fascinated with how much of an impact data can have on advertising. I don’t remember the exact technical terms that Brian used but I was impressed with how (TapCommerce) could identify each phone separately to each individual user and provide targeted advertisements based on user’s preferences.
-
The major takeaway I took from Brian’s talk on Thursday was how valuable big data is in today’s age, especially because of the rapidly increasing quantity. Not only were companies willing to pay top dollar to acquire personal information of individuals, but they later use that information to aggressively advertise to them. They view an individual’s recent web activity and then cater advertisements to them.
-
One take a way I got is how the big major companies are really looking for different ways to find new data about their company. There is a major opportunity for young people in this market to create a company with an idea like Mark to help these companies out and make a few dollars while your at it. The thing that shocked me the most about his talk is when he first got on their and I realized how young he was. It shows that you can access this market at any age really.
-
The most important aspect of Brian’s talk, and what I found to be most relevant to me, was the way companies are able to target their consumers very easily and quickly through the Internet. Social media plays a huge role in our lives and now it can play an even bigger one – helping us determine what products to buy. This feature not only innovates advertising, but what products people purchase. Consumers are exposed to new products they may have not been familiar with before, then they are only exposed to those products based on their purchase history.
-
I think it was very interesting to hear from someone who works so closely with a major company like Twitter. Big data was very interesting because since it is everywhere around us the information we don’t receive directly gets put online and influences the different search engines and the different ads. Consumers are being exposed to the new products online, on ads, etc. but they not be familiar with it but are interested in buying the product because they just tell that the product is great and so many people have used it. I feel like they should put a data graph on there website or ads showing how many people are interested and how many aren’t.
-
I think Brian’s talk about his first startup was interesting. I learned a lot about big data in a real life situations, how valuable big data is in today, how the big major companies are seriously looking for various ways to find new data about their companies, how companies use big data to reach the target ads based on a user’s latest activity.
-
- Load More
An example of a KPI is the measurement of a hockey player’s Corsi. Corsi is the shot differential for NHL players while they are on the ice. Basically shots for versus shots against. I use this very often to help with setting my fantasy hockey lineup because if a player has a higher corsi they are more likely to get shots on goal each night which then in return gets points for my team. Sports teams are starting to use a lot of advanced stats to help them determine who to draft or sign in free agency.
The amount of money I spend:
Specific: how much I spend daily
Measurable: the amount i make compared to the amount i spend
Achievable: keep track of money; be aware and reduce spending
Relevant: save more money, won’t spend more than i earn
Time: average money spent per week compared to money earned per week.
I use Google Analytics to measure traffic to websites. This is specific – measures the amount of views that occur over a time period. Measurable – shows peak times that the website is viewed. Achievable – shows the click through rate in which an individual goes from one page to the next. Relevant – shows what is pages individuals spend the most time and can gauge that certain topics are more important to readers than others. Time – I am able to see the average time spent on each page.
Class performance for a semester:
Specific: How well I perform in my courses over the period of a semester
Measurable: Grade point average
Achievable: Factoring the rigor of classes, extracurriculars, and past semester performance
Relevant: Based on how I need to perform to pass classes
Time: Over a period of a semester
Activity level per day:
Specific: how many steps in a 24 hour period
Measurable: a fit bit tracks my step count from the time I wake up – to bed time
Achievable: tracker will allow me to adjust activity level based on step count
Relevant: allows me to track my health and fitness day-to-day and see long term progress
Time based: tracker runs on a 24 hour cycle, time-locked day by day
Lead Conversion Rate
Specific: How many customers qualified (contacted and demoed) in a quarterly period.
Measurable: Number of customers that have sat through a software demo.
Achievable: Shows ratio of qualified to unqualified/contacted leads.
Relevant: Allows me to estimate pipeline and sales totals.
Time: Quarterly limits for measurable and reliable estimates.
Average number of steps via Fitbit
Specific: Steps in a given period
Measurable: Steps recorded in 24 hour period
Achievable: Can move less or more to alter number
Relevant: Gives deeper insight for those trying to track fitness
Time: Can check the amount recorded in 24 hours as well as past days
Working outs over a period of time
Specific: How much weight you lose at the gym
Measurable: Number of pounds or calories
Achievable: Can do more or less work to reach specific results
Relevant: Uses the number to see what else needs to be done
Time: Over a specific period of a semester
Heart rate during exercise on fitbit/apple watch
Specific: beats per minute while exercising
Measurable: resting heart rate vs. heart rate while exercising.
Achievable: keep track of beats per minute, does it change
Relevant: exercise more, lower your beats per minute while working out
Time: Minutes spent inside target heart rate during exercise.
A soccer forward’s performance level for a season
Specific: The player’s statistics determine how good/bad they are playing
Measurable: How many goals they score per season
Achievable: They can do more or less practicing to achieve a different set of statistics
Relevant: Allows the coach to determine whether to play this athlete or another one who is performing better
Time: Can compare the amount of minutes played to the amount of stats achieved throughout a given season or per week/month.
Data Usage
Specific: It keeps track of how much of a person’s cellular data is used up
Measurable: There is a concrete number
Achievable: A person can alter the amount of data they use when on their phone to prevent from going over.
Relevant: Most everyone uses data and it is very expensive
Time: Data usage is normally on a monthly basis.
Activity Level
Specific: how many miles walked/ran in a day
Measurable: a fit bit watch calculates how many miles are walked throughout the day
Achievable: can walk more or less to meet certain goals
Relevant: allows one to calculate fitness to contribute to health well-being
Time: fit bit runs day by day (24 hours)
How busy am I?
Specific: How often am I doing work
Measurable: Hours spent in a day not on free time
Achievable: I log the hours from when I wake up to when I go to sleep, I subtract the amount of free time I had that day (for a month)
Relevant: Allows one to be more productive with their time
Time: This can be done every day for at least a month to as long as one’s whole life, but a couple months seems like the best measurement.
Giving a tour to business partner over a period of time
Specific: How much time does it take to give a tour of a high school
Measureable: number of miles and steps to walk
Achievable: Can complete the tour in 10 minutes or 30 minutes
Relevant: Uses an agenda to see which classrooms they have to attend
Time: Over a specific time period to complete the tour
A KPI use on a daily/weekly basis is how many meal swipes i use on average
Specific: I have set number i can use each week
Measurable: I must budget them throughout the week
Achievable: To use all of them so I get my moneys worth
Relevant: Keeping on track so I do not run out by the end of the week and end u using cash
Time: Week by Week basic
An example of a KPI is how much money you spend.
Specific- How much money spent
Measurable- Amount of money spend compared to what is in your bank account
Achievable- Save money by reducing spending or by putting it in savings
Relevant- Don’t spend more than you have in your bank account
Time-Variant- How much money you spend vs. how much you have in your account. Also how much money you saved vs. how much you spent.
KPI: How many times I curse a day.
Specific: show exactly the frequency of me swearing
Measurable: count the number of time I curse
Achievable: just count the number each time I swear
Relevant: As many curses in one day, it shows that I probably have a bad day, lots of bad thing may happen to me.
Time: can count daily
My time spent procrastinating.
Specific – Shows how much time I spend not doing what I should be.
Measurable – I can count the amount of hours that pass before I get started.
Achievable – I can reduce the amount of time spent procrastinating by looking at the sheer stats of time spent being unfocused.
Relevant – The other aspects of life are affected by time spent doing nothing productive.
Time – Counted daily or weekly depending on tasks left to do
A KPI use on a daily/weekly basis is how much time it takes me to commute to and from school.
Specific: The amount of time it takes me to commute to and from school.
Measurable: I can count how much time it takes me to commute each day/week/month.
Achievable: I can adjust my schedule to take express trains or avoid peak hours to decrease my commute time or drive to school instead.
Relevant: Keeping track of the amount of time I spend commuting, that can be saved to complete other tasks.
Time: Counted daily or weekly depending on how often i commute.
An example of a KPI would be the word count when writing an article. It is specific because I either met the word count or I didn’t. It is specific because it is exactly how many words I have written. It’s achievable because I can figure out how long I should talk about one point in relation to the overall piece. It is relevant because I can know how much further I have to go until my task is finished. It is time-phased because I can figure out how long it takes me to write a certain amount of words and adjust my schedule accordingly.
How many times I eat
Specific:The amount of times i eat a day
Measurable:I count the amount of meals i have a day
Achievable: I decide the amount of meals i should have per day
Relevant: it can help me balance a healthy diet
Time:tracking by day
A KPI I use on a weekly basis would be the hours I’m at work.
Specific: Exact amount of hours I spend on the job at a weekly basis
Measurable: I can access my hour logs via temples portal and the KRONOS system
Achievable: I may come in later or leave early depending on my hour goals and workload.
Relevant: How much time I spend at work can effect my study habits or help me plan for my financial budgeting.
Time: Hours recapped on a weekly basis and logged on a daily basis.
An example of KPI is battery life on your computer.
Specific: It gives you the exact percentage of your battery life.
Measurable: It records it as you use it and as you charge it depending on usage or time plugged in.
Achievable: The less you use your computer the higher it will be, as well as when you charge it.
Relevant: It lets you know when you should charge your computer or when its going to shut off.
Time: It is refreshed all the time as long as you are on your computer.
A KPI that I use daily is how many calories I consumed in a day, that I calculate through a meal-logging app.
This is SPECIFIC and MEASURABLE because calories can be counted, and the app lets me know how many calories I am eating per item, meal and day.
It is ACHIEVABLE because based on my caloric intake, I can change my diet in order to consume less or more calories as needed.
It is RELEVANT for me to stay in shape and for my health.
It is TIME-RELEVANT because I can compare calorie intakes between meals, days and weeks.
My grades are an example of a KPI. The percentage value of my scores is specific. Every time I submit a new assignment or do well on an exam, the changes are reflected on my percentage; this is a measurable attribute. If I decide to spend more time studying for an exam or improving the quality of an assignment, the increase in my percentage value is achievable. Grade percentage is relevant for GPA and for any advancement purposes. Time is also a factor because it is limited. I do not have forever to work on an assignment, my goal is to give my best in the given time frame.