-
Shana Pote wrote a new post on the site MIS 0855: Data Science Spring 2016 8 years, 7 months ago
Some quick instructions:
You must complete the quiz by the start of class on February 22, 2016. The quiz is based on the readings for the whole week.
When you click on the link, you may see a Google sign in […] -
Shana Pote wrote a new post on the site MIS 0855: Data Science Spring 2016 8 years, 7 months ago
Here is the exercise.
And here is the graphic file you’ll need: Philadelphia Area Obesity Rates.png.
If you don’t have a local saved copy of the Milk vs. Soda exercise (4.2 – Getting Familiar with Tab […]
-
Shana Pote wrote a new post on the site MIS 0855: Data Science Spring 2016 8 years, 8 months ago
Here is the exercise.
Before you start, save this Tableau file and the studentloans2013 Excel workbook to your computer.
Remember, to download a file, click on the file link above, which takes you to the […]
-
Shana Pote wrote a new post on the site MIS 0855: Data Science Spring 2016 8 years, 8 months ago
Here is the study guide for the first exam.
Wednesday, Feb. 17 is exam review. (If we are current on material, we may have some time Monday, Feb. 15 as well.)
Format for review is:
Unstructured, […]
-
Shana Pote wrote a new post on the site MIS 0855: Data Science Spring 2016 8 years, 8 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 download the Excel file to do this exercise. To dow […]
-
Shana Pote wrote a new post on the site MIS 0855: Data Science Spring 2016 8 years, 8 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 […]
-
Shana Pote wrote a new post on the site MIS 0855: Data Science Spring 2016 8 years, 8 months ago
Leave your response as a comment on this post by the beginning of class on February 10, 2016. Remember, it only needs to be three or four sentences. For these weekly questions, I’m mainly interested in your op […]
-
Shana Pote wrote a new post on the site MIS 0855: Data Science Spring 2016 8 years, 8 months ago
Here is the exercise. And in Word format.
Remember to leave a comment on this post with the link to your graphic for our discussion.
-
Shana Pote wrote a new post on the site MIS 0855: Data Science Spring 2016 8 years, 8 months ago
In preparation for tomorrow’s activity on good and bad visualizations, you might enjoy this talk on the topic. Known as the “king of infographics” David McCandless is a renowned author and designer of spectacular […]
-
Shana Pote wrote a new post on the site MIS 0855: Data Science Spring 2016 8 years, 8 months ago
Some quick instructions:
You must complete the quiz by the start of class on February 8, 2016. The quiz is based on the readings for the whole week.
When you click on the link, you may see a Google sign in […] -
Shana Pote wrote a new post on the site MIS 0855: Data Science Spring 2016 8 years, 8 months ago
Here is the exercise.
-
Shana Pote wrote a new post on the site MIS 0855: Data Science Spring 2016 8 years, 8 months ago
This week we discussed “discovering” relationships in data that weren’t really meaningful (spurious correlations). There is a site dedicated to this called Spurious Correlations.
You can scroll down to the bott […]
-
Shana Pote wrote a new post on the site MIS 0855: Data Science Spring 2016 8 years, 8 months ago
Leave your response as a comment on this post by the beginning of class on February 3, 2016. Remember, it only needs to be three or four sentences. For these weekly questions, I’m mainly interested in your op […]
-
I think the most important takeaway from this week’s discussion has to do with the idea of bias. Usually in the cases of restaurants, the reviews are usually bad on the websites because the person felt so strongly that they decided to go and write a review. For instance, last week I went to a good amount of restaurants for restaurant week in center city. Before I went to each place I tried to get a general idea of it by reading the reviews and looking at the pictures. For one, the reviews did not do it justice. The place in my opinion was fantastic, but the reviews did not reflect it. I usually feel that people who post bad reviews expect a lot, which in turn causes them to never be satisfied. The only way to really counteract this problem is by posting a good review at a place if you had a good experience. This will help to balance out the reviews and try to make it fair.
-
The most important takeaway from last week’s discussion was how skewed marketing ads can be. We learned about how a lot of data is bias or is just a sample representation of what really is occurring. So it makes me wonder how much of what I see on billboards and TV is accurate and precise. I am a telemarketer and everyday I use data to determine what territory I am going to call. I may look at a lead sheet and scan over the customer’s name, residence, and list of referrals and based on that data I will decide rather or not I will call them. I do have a bias favoring the Virginia and Maryland areas when I am calling. This did not cause a signal problem for me and it hardly ever does because the data I use is backed pretty strongly by my results and percentages.
-
I think the most important takeaway is being bias. Some people just write bad reviews just because they have a problem the first time they go there. They do not give the places a second chance. For example, last week I went to Dave and Buster;s with some of my friends. Before I went i looked at a few of the reviews some customers have posted. Most of the reviews the people wrote said some bad things. However, when I went with my friends, it was amazing. The employees were nice and they got everything we asked for. The reviews did not do that place any justice. We had an amazing night and I would have given that an good review.
-
In my opinion, the most important takeaway from bias and the Signal Problem was that bias can expose you to only certain information, not all of the information available and can make you make your decision incorrectly. For example, If I wanted to buy a video game and I went to look at the reviews to see if it was actually a game I wanted to get or not. Bias could be the genre of the game because some people might review sports game higher than action games or vice versa and that’s what I would see on the reviews or my preference can get in the way of making a clear decision. This did result in a signal problem because more people favoring action games could’ve gave bad reviews on a sports game and I would’ve seen so many bad reviews that I could’ve thought the a lot people dislike it when it’s not a clear representation of all people reviewing this game. One way I could have counteracted this bias is to go on different websites which promote games I wouldn’t normally see and that could expose to information I wouldn’t normally have seen.
-
The most interesting thing to me was how bias is every where. Bias can lead to you only receiving certain information or even getting the wrong information. For example, my roommates and I wanted to go out to dinner and to a place none of us have been to. We all looked up the reviews and there were so many different reviews that you could tell some were definitely bias. Everyone has different experiences and opinions. With reading the reviews we trusted the people who had made multiple posts and seemed reliable, not the people who had only posted once because they felt so strongly about their experience.
-
Personally, I think that bias within data is much more harmful in the long run, especially considering that a bias in data isn’t considered in everyday activities. Recently, I’ve been looking for a pair of boots with very specific qualities: short, flat heel, and a light brown suede. I always use the same browser, and stick to the same few retail websites. As my search for these boots have continued, I have seen an increase in ads for short boots as well as for the stores I have been searching. Not until today did I realize how prominent the Signal Problem is: I was searching once again for boots under $150, and the results of my search were sorted by featured items. However, these featured items all fit within my previous search terms. Although this is not a negative issue, the same bias and signal issues can be easily transferred to more large scale problems.
-
I personal very carefully about the information I need to use to make decision. I cannot think of any decision that influence by bias, but I believe it does influence me. It is possible that I just did not notices. I believe bias exist everything and it influence a lot of people’s life. I will pay more attention into it. However, Signal Problem is very annoying. Every time I brought or view something online, it shows up many similar ADs about those items when I view other websites. It is just so annoying, and I could not figure out some way to avoid it.
-
I think the most important thing about last discussion is how easy it is to fake trustworthiness. It might take a while, but when just putting up a profile photo that might not even be real makes the user/reviewer’s words a little more reliable, there is no telling what a dedicated liar can accomplish. Of course, most people don’t have the time nor effort to pull off such scams, but it’s a scary thought nevertheless. As for a recent example, just a few days ago I was looking for datasets for the assignment. I have noticed that I kept looking for data in a few cities I know like New York, Columbus or California to compare with Philadelphia. In addition, because I like living in Philadelphia, I usually look for information that confirms my beliefs and ignores datasets that has Philadelphia being worse off than some other places. I only fixed this problem by looking back after I was done, changing my hypotheses to a fair mix of positive and negative opinions, and look for corresponding datasets.
-
In my opinion, the most important takeaway from the discussion was that you truly need to look into the specifics of the data. Draw your own conclusions about whether or not the data is reliable, because there is a large amount of unreliable data out there. Prime example is when I went to the fish store yesterday to purchase pet fish. I went to this large aquatic pet superstore that had a great selection. However, when I searched it I realized that it had a 2.5/5 star rating, and that cautioned me. After looking into the reviews, I had realized that the bad ones were quite biased based off of dumber complaints that people had. Issues such as overpriced food caused for a 1 star review, which I agree isn’t necessarily great, but is 1 star really necessary? Therefore, I went to the store anyway, and had a great experience! Awesome selection and the employees were very helpful. Always remember to truly look into the data, and try and figure out the true reason that it is presented in that manner.
-
In my opinion, the most interesting topic from the discussion last week was how often bias can be found in reviews. After searching for reviews about a couple of different places on Yelp (such as Temple, restaurants, and my previous place of employment) I’ve noticed that there was many reviews that had either a one star review or a five star review, with there rarely ever being any reviews with a rating in between. To add on to that, many of those extremely one sided views came from people who have barely done and reviews before, and for some it was even their only one. This made me realize that those reviews can’t necessarily be very trustworthy because one person may have been judging their opinion based on only one visit to a location. In fact, I even saw a one star rating for Temple on Yelp from a student who was angered because classes weren’t cancelled last Monday due to the snowstorm. The talk about bias has definitely made me pay close attention to reviews and make sure I only take them with a grain of salt rather unless there is a consistent number of reviews saying similar things.
-
I think the most important takeaway was the institution of bias. Personally, it would take the absolute worst and best experiences imaginable for me to write a review about a place. If I did so, it would be extremely biased, either staunchly in favor or in opposition of the restaurant. Most reviews of restaurants are similar, biased towards favor or opposition. My opinion/review would not be very trustworthy because my experience at the restaurant is exclusive to me. Thus, a lot of reviews should be taken with a grain of salt, because you are not guaranteed the same bad or good experience as others.
-
I believe the most important takeaway from the discussion was identifying trustworthy data. On sites like Yelp, many people are going to be biased one way or another whether it be for a school, restaurant, or retail store. I find the reviews that explore both the good and bad of a place the most helpful. A rating that is in the middle rather than towards an extreme is a review I am likely to trust more. Recently I have been searching for a place to live next year, and I have been reviewing each complex on whoseyourlandlord.com. Many reviews on this site are extreme opinions, more so negative comments than positive. I am using to this data to understand the positives and negatives of living in each of these complexes before I make a decision.
-
I think the most important takeaway is the personalization of information, due to signal problems and biases. Recently, I used data and reviews, found on google and yelp, to find nearby jiu jitsu training facilities. I have a bias because at home I trained at a Gracie Jiu Jitsu facility, and as soon I saw a Gracie facility nearby campus, I stopped looking at other facilities. The signal problem I could have encountered would be the elimination of other types of martial arts from my search e.g. boxing or wrestling. My interest is mainly jiu jitsu, for example on YouTube my recommended videos are solely jiu jitsu based. I could have counter-acted my bias by searching for other martial arts. Also, to counteract the signal problem, I should have erased all my internet history and broadened my search terms.
-
I think the most important takeaway is the personalization of information, due to signal problems and biases. Recently, I used data and reviews, found on google and yelp, to find nearby jiu jitsu training facilities. I had a bias because at home I trained at a Gracie Jiu Jitsu facility, and as soon I saw a Gracie facility nearby campus, I stopped looking at the other facilities. The signal problem I could have encountered would be the elimination of other types of martial arts from my search e.g. boxing or wrestling. My interest is mainly jiu jitsu, for example on YouTube my recommended videos are solely jiu jitsu based. I could have counter-acted my bias by searching for other martial arts. Also, to counteract the signal problem, I should have erased all my internet history and broadened my search terms.
-
I think the most important take-away from last week’s discussion is that you cant necessarily believe that the information you come across is completely accurate. Especially in the case of the Internet, where people are free to post/edit – there are bias’s present. I recently used the Google reviews and ratings when looking up a restaurant. I realized that I simply look at the star rating along with the number of “$” indicating the restaurant’s cost, whether it is cheap, moderate, or expensive. I never really took into account that what may be inexpensive to someone may be expensive to another. I also just took anything above a 4 star rating as a good restaurant, and was disappointed when I disagreed. I did not take the time to look at the number of reviews – one 4.5 star review could end up being the owner of the restaurant, which would have indicated, to me anyway, that the restaurant is really good. Next time, I will likely examine the number of reviews and maybe even read a few before making a decision based on the metadata summarized on the Google search page when looking up a restaurant.
-
In my opinion, the most important takeaway from the discussion last week was that not all data is entirely accurate. For example, reviews are loaded with biased data and can’t be credited as trustworthy, reliable data forms. It seems as though most reviews on Yelp and Google are extreme opinions, where individuals post either very negative or positive reviews, and give extremely high or low ratings. These reviews are based off of individual’s personal opinions and experiences and it seems as though only people with extreme opinions take the time to write reviews on these websites. It is very rare to find reviews and ratings that are in the middle, and I personally believe these reviews are most credible. The next time I look to buy something online, I will be certain to read a couple reviews to decide whether or not the data is reliable and not just simply filled with extreme ratings and reviews.
-
Last week’s discussion brought up many good points about data and how a lot it can display a bias. For example, google has an option to review a place, and one review for Temple University was rated one star. The person who posted the review was very upset about how Temple University students had class on a very snowy day, and that caused him to leave a bad review. One bad experience at Temple caused him to say that Temple University is unsafe and he doesnt feel comfortable having loved ones go here, which is completely untrue.
-
I found last week’s readings and discussions very interesting, especially the talk of the different bias in data. This is very important when using data because there are many underlying factors not even considered that could have a significant effect on data. Last week, when my girlfriend and I were deciding where to eat, we took to yelp to decide. This can be a fantastic resource but it also can be filled with hidden bias. It is very possible accounts are fake and are used for business owners to laude what’s theirs and to manipulate reviews for competition. Realizing this, we only paid attention to profiles that seemed credibe.
-
The biggest point that stood out to me this past week was that data can be very untrustworthy in certain situations. our discussion on the Yelp and Google reviews proved that to me. Last week during restaurant week I was picking a place to go and eat. Instead of using the reviews on yelp I asked a few close friends from around the area about a few different restaurants and got their opinions on the matter because I knew that they would be unbiased and truthful in telling me which restaurant to go to. I also got more in depth with them about what they ordered and suggested. Their reviews could also have a certain bias but I trust them more than I trust the Yelp reviews.
-
The most important strategy that I took away from our class discussion on the Signal Problem was being able to differentiate the truthful reviews from the bias reviews. About a week ago I was looking online for a place near by to get my haircut. I found Diamond Cutz and looked though a few reviews. Every single review I looked at was very positive except one. The customer gave a poor rating and no explanation as to why they gave the bad review. I took this as a bias review since it was the only one of dozens and went to Diamond Cutz anyway and it turned out to be a great place!
-
The most significant knowledge that i gained from last week is that not all data can be trusted reliable data and not all data correlates with each other. In some case studies a lot of information is collected but not all of the data is useful to solve the problem at hand. Signal also pays a key factor in what data and reviews are valid as well because we have to take into consideration who is delivering the data. Websites like yelp and amazon can sometimes give falsified data because someones bias could have a profound effect on their review of the product or service.
-
I believe that both signal problem and bias were two very interesting topics that we covered last week. It’s very evident that individuals can create a signal problem (especially with review based sites) as they are generally going to give an extremely bias review based on a possibly good or bad experience they may have had; if everyone instead were to review no matter their experiences we could have much more reliable data. To counteract issues like these we need to find a way to incentivize people giving their reviews regardless of experience or bias.
-
The most important takeaway from the discussion for me was that you can’t escape bias, and that it’s better to get used to learning how to identify untrustworthy data that might stem from bias. Many reviews you see, whether they are for restaurants, products, services, etc., are comprised of either glowing reviews or nasty ones. It can be difficult to determine which of the “glowing” reviews are trustworthy and which were written by the company owner’s sister-in-law or great-nephew. It can be equally tricky to identify which of the “nasty” reviews were written by people who have a flair for the dramatic. I recently purchased a new laptop, but beforehand I researched extensively on reviews, consumer reports, etc. I definitely had a bias towards Macs, however, and found myself comparing everything against the laptop I really wanted. While I still took the reviews and reports seriously, I will admit that I was trying to research others just to confirm my decision to buy my first-choice Mac. Counteracting bias in general requires that we take everything with a grain of salt, and really the only way to be sure about most things is to try them out for ourselves (restaurants, products, and services).
-
In my opinion the most important aspect that I had got from last week’s discussion on bias and the Signal Problem, is that finding good quality and reliable data relies on firstly good information and also, while not always necessary but greatly appreciated, the effort and time spent from an individual in making and delivering the said data. Many times short lived, no-name reviews on restaurant reviews for example, doesn’t seem credible because of its illegitimate nature. However, when we see a full name, and a carefully explained review noting positives and negatives we tend to value that review over others, even though a bias will be present. A recent example on how I used data coupled with bias in my decision making was in choosing certain courses for Spring semester. Using the data on sites like RateMyProfessors.com sometimes prove accurate when choosing courses that certain professors teach, while at times they might not however my bias comes into play when I see certain reviews I like given what I mentioned above where an individual takes time to give a detailed explanation to how good or bad a professor is. Sometimes /I find the reviews completely contrary to what I read, causing a signal problem with the ratings/ or data that was accumulated for a professor. A way to counteract my bias in regards to this signal problem would be to give a detailed review myself or to purposely enter a class where the professors rating is low and see for myself if the data is an accurate representation.
-
During last weeks discussion on the bias and signal problem I think the most important takeaway was that most data is unreliable according to bias. Just from looking at the different reviews of Universities in our area I found that most reviews were from current students or alumni with only positive comments which is extremely biased. I have been noticing bias in reviews more after this discussion. My friends and I have been trying to find new restaurants to eat at but, looking at reviews first and mostly finding positive reviews featured first even if there are more negative reviews. this just makes me realize I must take everything into account because the internet filters my view of the world without even knowing it.
-
The most important takeaway from last weeks discussion was how bias can make many reviews unreliable. The reviews are often unreliable because many people write a really positive review or a really negative review based on one experience. You can’t really decide how a your experience at a restaurant will be based off another person’s only time at that restaurant. For example, over the weekend I decided to look at some reviews on yelp for restaurants in center city. Sure enough, there were numerous one or five star reviews about the restaurants. I chose to go to a restaurant that had a few one star reviews for different reasons to see if I experienced the same problems. I had no reason to complain because both the food and service were great. The best way to counteract a bias you may find on a review site like yelp, is to go to a place in person or ask someone who has been there multiple times.
-
I believe the biggest takeaway from last week’s discussion was the trust in data. Not all data received is necessarily correct and accurate via bias, that leaves us, the consumer, to try and fact-check the data if we are not sure if it is correct. Data i’ve used recently was purchasing a MacBook Pro. I did a lot of research about the product and its protection against viruses as my previous laptop succumbed to a violent virus resulting in my loss of all my personal documents. I checked many different reliable sources to figure out if a Mac was reliable and, according to all my sources, the MacBook was far more reliable in terms of virus protection and performance in comparison to my previous Dell laptop.
-
The most important part of last week’s discussion was bias. While reading through reviews of places to eat the best cheese-steak, bias was very apparent. The reviews were very extreme, either very supportive of how good the sandwich was or how poor the sandwich was. You never really know how good the cheese-steak is until you experience it for yourself. The best way to counteract a bias is to read reviews from people with profile pictures and people that may give 3-4 star reviews. Reviews that may seem long also might cover every aspect of what is being reviewed so these are more trustworthy.
-
I think one of the most important takeaways is that the facts we are finding on the internet may not always be “fact.” Due to our locations, preferences, past searches, and other factors, the information that we are searching for is tailored to what we ‘want’ to see, versus what may be more true or something else. I have had this happen to me when searching information for a race/ethnicity class – I do not identify as a political party, as I have varied opinions on different topics (most would probably consider me to be more liberal) and, when searching for facts about welfare and who receives them, I would receive statistics that would upset most “republican” ideologies about welfare.
-
In my opinion, the most important takeaway was that it is extremely important to make sure you know where your data is coming from, so you can identify any potential biases that could be associated with it. I recently made a decision to avoid purchasing a video game that I had been looking forward to playing. I used a review from a site called IGN to help me make my decision. In this situation, IGN actually did a good job of avoiding biases by hiring someone unrelated to the company to do the review, because one of IGN’s main reviewers had been involved in the game as a voice actor. Had an IGN employee written the review, they may have had a bias towards the game and given it a positive review that it did not deserve in order to help out their co-worker and give the game positive publicity.
-
The most important take away for me from the conversation we had on data biases and signal problems is that you have to take everything with a grain of salt. Do not believe everything you see because more data are usually needed. For example, when looking at Yelp one has to account for the user who posted, the amount of reviews the search has, and the content of the reviews. Recently, I used data from a annual report in order to formulate an idea on a company’s future progression. The bias here could be the fact that the company is the entity that wrote the annual report, therefore, I would have take biases into consideration.
-
I think the most important takeaway from last weeks class was the biases we face when collecting data. Particularly, we face many biases with data retrieved from online sources such as yelp. Yelp can be a very useful piece of data. However, the data found on yelp can be subject to many biases. We were talking in class about people’s emotions being a cause for bias on sites such as yelp. If you go to a restaurant after having a bad day and you also have a bad experience at the restaurant, maybe this was just the end to a really bad horrible day and your emotions made you a little on edge and therefore made the experience worse than it really was. That would cause a bias in your comment that others read when looking for data on that particular resaurtant you visited. This is why data that comes from sites like yelp can’t ever be depended on completely.
-
Along with everyone else, I agree that bias is a topic that I stuck with me from the lectures. Biases are everywhere and we cannot escape them for as long as we are living, but it’s biases that can lead to misleading or irrelevant information. For someone who is always on the move or always trying to look for new restaurants or articles to read I subject myself to data and biases every day. From the lecture, biases have taught me a powerful lesson…. question everything! I’m now looking back at all the times I would tell friends about new restaurants, places to hang, or articles I’ve read and how certain I was that the restaurant or place was so cool, or how much I learned from the articles and it is probably a 150/50 chance that some of information I interpreted was either false or not credible. This makes me aware of what I put out into the search engines and to look for more resources about similar topics if I want to really invest my time trying to go there or learn about it. Not that what I have read was pointless, but it skewed my judgment because I believed that was the only or popular source. Now I know it is better to have seven different perspectives than one.
-
I think the most important takeaway from last week’s discussion was that bias is everywhere in our lives. In last week’s class, we found bias on Yelp and in Google reviews, and we discussed how biased those reviews can be. In my opinion, bias are unavoidable, because in fact, reviews are very subjective ideas based on personal experiences. For example, I once went to a nice restaurant with a friend, but we argued during dinner, which really influenced my judgement, so if anyone asked me about that restaurant, I would give four stars though it might be worth five stars. Now I try to counteract the bias by only describing the facts. And although it is hard to eliminate or escape bias, I think as long as we are aware that most data are biased, we can drop the misleading data and get the “right” information.
-
The most important thing that I learned in last weeks discussion was that bias is opinion based. I know that sounds redundant but you have to remember that you are not the people reviewing this topic. Even though some people may feel very strong one way or the other, when there a multitude of opinions you get see a general consensus about a certain restaurant, game or show. Problem is is that you are not part of the general consensus and may have different tastes then the majority of the people reviewing the item. When i went to look up reviews for a new show that came out this season to see if it was worth picking up, the majority of reviews said the show was boring, slow and generic. Thankfully,the summary of the show was enough for me to give it a shot and after watching the first three episodes I thought the show to be perfectly paced for what it was aiming for and unique when compared to other shows in its sub-genre. The way I personally counteract bias is by reaching out and getting opinions from people like me, and if i can’t fin anyone like that to give me an opinion, try it myself.
-
I think that one of the most important takeaways from our discussion last week was the Filter Bubble. I never knew that they could/would do that to us! A decision I made recently that I had to research for was the type of drum set I wanted to get. I looked at reviews on the website as well as videos on Youtube so I could see how they sounded and judge for myself. A bias that became present on Youtube was the publishers of the content I was watching about the drumsets. The people who made the drum sets were posting the videos, which only involved good things about the drums. When reading on the internet about them, from forums to places where they were purchased, people had other opinions about certain features that were highlighted in the videos. I began to not pay as much attention to the videos and started focusing more on the written reviews due to the signal problem at hand. The way I counteracted the bias and Signal Problem for my case was that I watched videos of people playing the drum sets to songs rather than just playing to review them. To steer away from user bias due to their drum skills I watched various players on the same sets to ensure I found the right ones. Through this experience I learned that the best data speaks for itself. I also learned that less bias in data leads to more confident predictions and outcomes.
-
I think the most important takeaway from last week’s discussion was to watch out for bias reviews from online websites. It is important to keep in mind that a review may just be based off of one really good or one really bad experience at a certain place.Also, I learned that I should not base a decision off of just one review, I should read multiple reviews. For example, a few days ago I wanted to buy a product from a seller on Amazon, but first I wanted to read reviews on the seller to make sure he or she had a credible reputation and would send me the real product I was looking to buy. Some reviews were good and some were bad, but there was definitely a bias in the responses because most of the reviews were based off of just one interaction with the seller.
-
I think that the most important takeaway from last week’s discussion was that bias is everywhere in our lives. In last week’s we found that we can find reviews on either Yelp or Google review. In my opinion, bias are certain because reviews are very smart ideas based on personal experiences. For instance, before I go see a movie I will like to read other people’s comments to see if the movie is a good movies and if worth my money.
-
The most important take away I got from class last week was the idea that we must be more careful in looking for bias in the data that we choose to use. While looking at the reviews in class from Yelp and Google I came across a lot of sketchy reviews. When me and my friends were looking at reviews for this restaurant recently I was paying most attention to the negative reviews because I had previously had a bad experience myself which is showing bias. I could have had my friend look over the reviews knowing that she had never been to the establishment before.
-
The most important take away from last week’s class was that I got a good in site of filter bubble and how search engines and other popular websites track our activity to sell ads and make money. I also discovered that the enormous amount of bias on the reviewing websites creating signal problems. Before last week’s class I looked at review sites as completely positive and reliable. After actually analyzing popular sites like google and yelp I was able to figure out the real deal behind reviews. Now I can get a good sense of idea identifying biased reviews. We recently googled a good place to get a haircut nearby as my friend wanted to get a haircut. After looking up some places nearby and reading review about them on google, we were able to identify the false, misguiding and biased reviews. analyzing little things like total number of reviews and the patterns of the reviews, finally we were able to sort out the best option. This would have been impossible if we would have trusted all the reviews out there.
-
The most important takeaway from last weeks discussion would undoubtedly be the issue of bias when it comes to data. With this I specifically thought of how bias plays a huge role in me choosing my classes for the semester. I usually finalize my decision on whether or not i’ll remain registered for a class based on what I read about the professor on a rating website. I can honestly say that I trust the opinions of my fellow students enough to change the time and even the day of my class if the professor does not get a good rating. I depend on the data provided from that website highly every semester and would feel genuinely anxious about how my classes would pan out without me knowing a bit about the professor beforehand. There have been times when I’ve tried to overlook certain biases that I see within the information given on a teacher just because a certain class may fit flawlessly into my schedule, yet if it ends up being as bad as stated, i move on to the next class.
-
I think the most important takeaway from class is the importance of online reviews. I grew up in a time with the internet thus the ability to rate items online. An example is anytime I buy something online. I will always check reviews online. These biased opinionated reviews will always skew me making the purschase.
-
In my opinion, the most important takeaway from the class discussion is the topic of biases. Recently I have been trying to coordinate a birthday party for a close friend. I decided to use Yelp and other review sites to find a good restaurant to celebrate the cause. After reading many reviews i started to notice a bias trend that people either loved or hated the places they went too. This caused a problem in my decision making because the data was skewed. Inevitably I chose the restaurant with the most star reviews.
-
What I took away was the importance of realizing that signal problems exist. I never really though about it until you mentioned the example about hurricane report. It makes absolute sense. Data can be easily skewed due to many reasons so know the source and analyzing it careful is important. A recent example of how I’ve used data to make a decision is when I decided to go Vegan. After watching a few documentaries on Netflix and being exposed to the data on the environmental impact our daily eating habits has I made a decision to stop eating meat and dairy. The data they emphasized in Conspiracy may potentially be biased since they were trying to persuade their viewers in one form or another. Because the numbers where so large (e.g they mentioned that it takes 1100 gallons of water to come up with the average portion of beef an american eat daily) I actually did my own research and found that they weren’t far off. This being said I think bias and signal problem in data is really important point to factor in when analyzing data from a large source.
-
-
Shana Pote wrote a new post on the site MIS 0855: Data Science Spring 2016 8 years, 8 months ago
As emailed, today’s session was recorded via Class Capture for anyone unable to make it to class due to snow, ice and travel conditions.
You can view the recording here.
To receive full credit for attendance […]
-
Shana Pote wrote a new post on the site MIS 0855: Data Science Spring 2016 8 years, 8 months ago
Some quick instructions:
You must complete the quiz by the start of class on February 1, 2016. The quiz is based on the readings for the whole week.
When you click on the link, you may see a Google sign in […] -
Shana Pote wrote a new post on the site MIS 0855: Data Science Spring 2016 8 years, 8 months ago
Here is the exercise.
As discussed in class today, please comment on this post with the following:
What dataset did you find
Where did you find it
Why did you think it was interesting
What did you […] -
Shana Pote wrote a new post on the site MIS 0855: Data Science Spring 2016 8 years, 8 months ago
Reminder, as discussed at end of class yesterday, you’re to complete In-Class Exercise 2.1, Creating a Data Dictionary, and bring your results to class tomorrow for discussion.
In the spreadsheet linked from […]
-
Shana Pote wrote a new post on the site MIS 0855: Data Science Spring 2016 8 years, 8 months ago
Leave your response as a comment on this post by the beginning of class on January 27, 2016. Remember, it only needs to be three or four sentences. For these weekly questions, I’m mainly interested in your op […]
-
I found this article especially interesting since it has to do with something that most people don’t usually think of when they think of Big Data. In terms of growth, I think that data can help propel the legal field forward light years. Most of the time Lawyers spend their time preparing and researching cases, with the help of data and programs it could help make lawyers lives a lot easier and more efficient.
-
I found this article very interesting because many businesses use statistics and data to make profit. However, some of them are not confident enough to use the data they have gathered, so their businesses start to decline. They pay all of these data analysis people money to gather and analyze the data that they do not use. If they used the data the experts have gathered, then they will have a better shot a earning more money.
-
How Barclays is cashing in on big data & Hadoop to stay ahead in fintech
This article was interesting because it shows how banks can use big data to their benefit and get an edge over their competitors. For example, Barclays is beginning to realize that this specific data is “not only for the IT department but it can be used as a customer assets that can treat effectively as a business.” If other Banks and financial planners use data and big data to their advantage, then they can gain the advantage of their competitors and excel at growing.
-
http://www.wsj.com/articles/u-s-tech-companies-bring-encryption-battle-to-davos-1453320950
This article was interesting because the request for more data and more surveillance is causing tensions between different governments and technology companies. Governments want U.S. technology companies to give them more information, however the U.S. can’t really do that. These governments, such as the European government, want more data on people’s devices to help combat terrorism. It is difficult for the U.S. to give out more information due to a law that is in place, only limited personal information can be turned over to the other governments.
-
http://www.techrepublic.com/article/datas-new-home-your-companys-balance-sheet/
This article is interesting because it points out how data is now the new question mark for the accounting departments and corporations as a whole. Data is an asset that could amount to trillions in value; it can be depreciated, exchanged and used to make predictions and/or insight; and there is also a cost in maintenance and safekeeping of data. However, it’s hard to manage what you can’t measure; and so companies have to find a way to monetize the costs and benefits of data if they want to stay relevant in the era of digital business.
-
http://profusion.com/2016/01/14/how-data-will-shape-finance-in-2016/
I found interest in this article because it mentioned how some of the big banks are integrating Apple payment systems. I known Apple for their ipods, iphones, and ipads, but for them to be making a leap into the world of banking is pretty impressive. At the same time I feel like Apple should stay in its lane and continue to dominate the cellular industry. Another thing is that all Apple products run together and by them getting involved in banking it can only make you think that cyber theft and ID theft will become an easier task for even the average person to do. -
I found this article interesting because it connected what we will be learning in this course with my accounting major. The article explains how Big Data can enhance an accountant’s skills. I now know ways that Data Science will be beneficial to me in my career path. -
http://www.bloomberg.com/features/2016-millionaire-odds/
This was in interesting article from the Bloomberg business report. It collected a lot of data on the basis of how likely you are to become a millionaire. It shows really cool graphs and charts visualizing the chance you’ll become a millionaire. It takes into consideration your age, race, and education level. I found this interesting and relate-able because, well, who doesn’t want to be a millionaire.
-
http://fivethirtyeight.com/features/us-markets-china/
This article, titled “US Markets Won’t Necessarily Follow China into the Abyss”, commented on the mini market meltdown of US Stock Markets the morning of January 20, 2016. It asked the question whether China’s volatile markets and economy caused the mini meltdown, and if this pattern would continue in the future. Recently, China’s stock markets have dropped over 40% since last June, and there is some correlation between the Shanghai Composite and the U.S. Stock Market. However, this correlation has only been consistent since mid 2014.
-
http://fivethirtyeight.com/features/the-most-ominous-snow-forecasts-for-new-york-city-were-right/
Having lived in Manhattan for 7 years, this article caught my eye. It talks about and compares the city’s storm forecast for this weekend with that from last January when New York was supposed to get hit with almost 2 feet of snow. In reality, we barely got anything close to that amount last year, and I’m sure as a result this year many New Yorkers were far more skeptical. The article says how this year the forecasts gave such a wide range of possibilities, with actual data minimums and maximums for snowfall (from less than an inch to 23 inches), that it made it hard to know what to expect. Ironically, the city is close to hitting the forecasters’ maximum prediction of 23 inches (whereas last year it only snowed a few inches, contradicting forecasts of over a foot and half). -
http://fivethirtyeight.com/features/fivethirtyeights-guide-to-predicting-the-oscars/
All movie watchers love watching the oscars to see which of their favorite actors will win film making’s biggest award. FiveThirtyEight.com developed an “election model” to predict which actors and actresses are most likely to win an award for the category that they are nominated. Their model is created by taking the data from previous awards that actors/actresses have won. They then test to see if that award has any correlation with who will win the Oscar. Depending on the award they’ve won, they are given points based on how likely the winner of that award is to win an Oscar. The person or movie with the most points will be most likely to win an award at the Oscars. -
http://news.mit.edu/2016/quantum-approach-big-data-0125
This article is interesting because it addresses some of the problems with analyzing data that we have discussed in class. MIT has developed an algorithm that assesses data based on topology, which means that data collected, such as the amount of something or how many parts of it are connected, remains the same no matter what way the rest of the data can be stretched. The algorithm can be used to make decisions based on factors like these to minimize error in the way data is collected and analyzed.
-
http://fivethirtyeight.com/features/why-isnt-anyone-talking-about-the-deficit-anymore/
I am currently a finance major and very interested in politics. After reading this article I realized that the federal deficit is not debated as much as the previous presidential campaign. The data demonstrates the steady decline in interest of the deficit, however I believe it is still a major issue that deserves more focus than it is receiving at the moment. I am closely following the primaries and enjoy listening to the candidates debate many topics; however, I cannot recall one deficit spending or debt related question appear in any of the debates I have watched.
-
What sparked my interest in regards to the subject matter on this article is how it closely relates to the discussions we were having in class on privacy and personal information being put online for both public and private personal use. As the article points out many internet users choose how they share their personal information based on the context in which it is used, often using their discretion on how much they share based on the benefits that they yield from the type of service that is provided using their personal information. These range from services such as sharing health information which is vital to street surveillance camera’s which depending on the individual would deem the use of their personal information acceptable or not. -
http://memeburn.com/2016/01/five-big-data-trends-that-will-impact-south-africa-in-2016/ Five big data trends that will impact South Africa in 2016
I am a MIS major. This is my first MIS class, so I really do not know much about it. However, I always be excited about information and data. Big data is now being offered at universities in South Africa, and I am so happy for them. This article also mentions that we are a part of it and we can make money from it. As a business major, I agree that this article said “Big Data to truly understand their customers in order to offer them tailor made solutions.” This is so true.
-
An Italian newspaper and 18 websites owned by the same publisher published a series revealing multiple assets that were stolen from the Italian citizens by the Mafia. Titled “Confiscati Bene,” or “Well Confiscated,” the series was completed in three steps. The first was scraping data from an agency database that held a records of all the stolen assets. They then extracted and curated their content from “some 3,000 pages of documents.” After “Confiscati Bene” was published, it was uploaded to a data catalog and the data continues to be updated.
-
This article is of interest to me because it discusses the percentages of young college graduates who are currently working. As a college student, this data is very helpful and really gives students an idea of how important a degree is in today’s world. In the U.S. right now, about 80% of young college graduates are working. Finally, this data goes to show us that earning a degree doesn’t guarantee anything. I will continue to follow the data concerning these percentages and hope the amount of graduates working increases as I grow closer to graduation.
-
http://www.providencejournal.com/article/20160114/NEWS/160119475
I am currently an undeclared business major, but I am particularly interested in the long-term effects that concussions have on individuals. The data driven analysis in this article relates to myself personally because of my history with concussions. I have suffered 5 concussions and frequently research articles similar to this. This article demonstrates data based on the amount of young kids and teens who are suffering from concussions, the physical shape they are in, and how they recover. It also explains the patterns of sports related concussion injuries. -
http://fivethirtyeight.com/features/spurs-warriors-is-the-best-defense-offense-clash-in-nba-history/
This article is about the Golden State Warriors and San Antonio Spurs basketball game tonight. I am a huge basketball fan so reading this wasn’t even like homework. The two teams are off to a historic start and very rarely does one see such teams play during the regular season. The article from fivethirtyeight dove into the data behind it which made it even more interesting. It went all throughout the history of the league and compared the statistics of great teams playing against each other. It’s prediction is the better defense usually comes out on top that would be the Spurs then.
-
http://fivethirtyeight.com/features/the-stock-market-is-not-the-economy/
This article is titled “The Stock Market is not the Economy”.
I liked a few things about this article. First off the overall picture, which a lot of people tend to feel otherwise, is that the stock market truly isnt the economy. I am an economics major, and finance is just a sector of economics and many people seem to not understand that. I am not saying that it doesn’t play a role to the health of the economy, because it does. However there are many other factors in play as well. Also, I really enjoyed the wealth inequality statistics that were in the article. When you research this data (which I have done before), it tends to be misleading and somewhat biased to the feelings of the purpose publishing the data. A great example from the article is the data which he states “Americans have more money, but also more debt.”
-
http://fivethirtyeight.com/features/are-we-headed-for-another-recession/
I found this article interesting because it was seeming that the economy was starting to patch up a bit. This article proves other wise stating how the Stock Market had the worst start of all time. Our economy only seems strong on the surface. Showing how big data like what Obama said in the State of the Union talking about all the upsides of economy he is working toward that when further researched may call for the near down fall. This is not necessarily related to my major but, if there was another recession my major would not be top of priority job wise. Just having a job during a down time such as a recession is as good as it could get. -
http://www.forbes.com/sites/ciocentral/2012/07/05/best-practices-for-managing-big-data/#41a2d895ef02
I found this article interesting because many business use data to make profit. However, some of them are not familiar with the data they have gathered, and that makes their business decline. If they use the data they will earn more money.
-
http://www.espncricinfo.com/magazine/content/story/964643.html
I find this article very interesting because this article shows what should happen If anyone leads a side for a long long time. Readers can found in this article the performance and captaincy skill of M S dhoni of his early years of captaincy and now. Ian Chappel has nicely stated why the indian cricket team is fail to find the success in the recent past under Dhoni’s captaincy. He also give the solution to BCCI to get rid of current bad patch.
-
http://www.forbes.com/sites/quora/2016/01/21/how-urbanspoon-collects-data/#21f549612b5d
I found this article interesting because it introduces my favorite mobile app, Urban Spoon. I’ve been using Urban Spoon for months, and I’ve been wondering the way it collects data and the difference between the data sources of Urban Spoon and the data sources of Yelp. After reading this article, I found Urban Spoon uses data from users, magazines and newspapers, while Yelp only collects data from users’ reviews. I think adding data of public media to the database is really a smart move in the rating app market. -
http://fivethirtyeight.com/features/billion-dollar-powerball-lottery/
I chose this article about the lottery because at my part-time job I am a lottery clerk. I witnessed first hand the Powerball craze when the jackpot hit one billion dollars. In addition, I am an accounting major, so naturally I am interested in numbers, and the few sentences that it mentioned the after-tax jackpot. I found the statistics about the Powerball interesting. I was curious as to how many tickets were sold. I found some of the things that the article stated were shocking. Specifically, where it suggested not to play the Powerball with the intention of making money. I found this a little funny because most people play the lottery to try to get rich.
-
http://blogs.wsj.com/digits/2016/01/26/waze-to-scoop-up-more-data-through-new-partnerships/
This article is about how waze is partnering with other companies to help improve their data. They are partnering with a company that is meant for ambulances in order to keep people updated as to where ambulances are, and keep ambulances updated as to where accidents and backups are. It’s interesting that even partnering with a couple of new companies can greatly improve and increase a companies data.
-
As a Tourism and Hospitality Management major and avid traveler, I always want to know I’m getting or giving a good deal when it comes to traveling, especially within hotels. At almost any hotel now, you can sign up for some type of rewards or loyalty program, which encourages guests to continue to come back to the company or specific hotel. In this analysis, loyalty programs are stacked against each other in an era when many hotel chains are merging and buying more products – this may cause some devaluation of the products and services provided by these hotel chains.
-
http://www.stats.com/data-feeds/
This site is great for anyone who is interested in any aspects of sports weather it is the stats about the players/games themselves, fantasy sports predictions, and predictions for the games possible outcomes. this website is great for people who love daily fantasy sports so they can construct the best possible team to win a large prize. -
This article is interesting to me, because although I’m currently an Undecided Business major, I’ve been leaning towards Accounting. It talks about different types of analytic tools that Accountants have been developing in order to help track data which in the long run will be able to help business firms and accountants maximize their profits. It discusses how businesses are able to gather data from consumers in order to take advantage of that information in order to reimagine their business in the long run. -
http://fivethirtyeight.com/features/the-panthers-first-half-blowouts-are-unprecedented/
This article discusses the success that the Carolina Panthers have had this season on their way the Super Bowl 50. This is interesting to me because I am a huge football fan and it was interesting to look at the data and their conclusions in this article. This article explains how the Panthers have managed to score a lot of points quickly during games. It shows that this Carolina Panthers football team has had the fastest playoff starts in NFL history. They’ve managed to end the first quarters of their playoff games with a 91% win probability. However, the article goes on to state that their success in the first two playoff games doesn’t have any correlation with success in the Super Bowl. The article also states that the top six teams based on winning percentage possibility in the playoffs have failed to win the Super Bowl. -
http://www.wsj.com/articles/outside-voices-davos-importance-to-marketers-grows-in-turbulent-times-1453407649
This article was interesting because I never really thought of how important data would be in marketing. Now that I actually sit here and think about it I can come up with a lot of ways it would be used but I just never paid any attention to that aspect. This article focused a lot on how companies will have to learn how to collaborate to help push forward with the marketing industry due to data and technology advances. Stating that no one can have the answers to all the problems which actually makes a lot of sense. -
I found this article interesting because I am an actuarial science major about to enter the insurance industry. This article summarizes the importance of data in the industry and emphasizes the need to be able to work with that data. I also found it interesting that a big insurance company like AIG actually has a “Chief Data Officer”. I think it just proves the growing role of data.
-
http://fivethirtyeight.com/features/a-day-in-the-life-of-an-iowa-family-drowning-in-campaign-ads/
As a marketing major, this article was very interesting to me. It gave me insight on how too much marketing and campaigning can effect society. Sometimes, all of the ads can get overwhelming and annoying. It is important to market in a way that does not make the consumers feel as if he or she is drowning in advertisements. Ads are everywhere. They are on TV, billboards, flyers, social networks and much more. Marketing professionals can easily not realize when they are”over marketing.” As a marketing professional, I will be mindful of how too much advertising can effect consumers. -
http://fivethirtyeight.com/features/december-jobs-report/
This article discusses the large job growth in December and whether or not it will last heading into 2016. Employers have added 2.7 million non-agricultural jobs in 2015 which was slightly less than in 2014, but still a great amount of jobs. Also, unemployment is at a very natural rate which would help these jobs last through 2016. Wages have increased by 2.5% which is also a good sign and will help those dealing with inflation. Overall, I believe the job growth will last through 2016. -
Article Published: January 25, 2016
Finance Major: Apple’s earnings
URL: http://blogs.ft.com/ftdata/2016/01/25/russia-gdp-iphone-sales-mmr-rates-and-a-history-of-bear-market-the-day-in-4-charts/
Why is this interesting for a finance major? The reason I chose this article is because I am interested in the stock market. By analyzing the data given from the video, I was able to conclude that Apple’s stock price may stabilize or decrease in the future. Apple’s sales never went on a decline for the first quarter since it started out and may show its first Q1 decline based on past data and an estimated 2016 fall data. -
http://fivethirtyeight.com/features/how-math-and-not-a-telescope-may-have-found-a-new-planet/
Two scientists made headlines after having reason to think they have evidence of a possible new planet. This “Planet Nine” is 5x the mass of Earth and in the densest part of the galaxy outside of the solar system. They believe this to be possible after noticing a number of icy masses moving/orbiting in the same direction, which is unusual without there being something to cause it to move that way. The two scientists created a mathematical model from the movements from the ice masses, which led them to believe “Planet Nine” may exist. I find this article interesting because I am intrigued by outer space and want to know what’s really out there in the galaxy.
-
http://www.forbes.com/sites/bernardmarr/2016/01/26/how-big-data-and-analytics-changing-hotels-and-the-hospitality-industry/#5021f0364b39
This article is interesting because it shows how big data is used to differentiate and value different types of guests in the hotel and hospitality industry. Being a business major and having an interest in real estate, this article has great insight on how the industry views its guests. The article also explains how big data can be use to help identify and value different types of customers and how to determine which one can add more revenue whether in the short term or the long term depending on the location of the hotel. -
http://www.scientificamerican.com/article/how-data-brokers-make-money-off-your-medical-records/
This article particularly interested me because It reminded me of the box you check when you have a doctors visit which basically states whether or not you are okay with that doctor’s office or hospital using your medical information anonymously for research purposes and such. The article highlighted this as it explained how certain companies obtain medical data from hospitals and medical practices and organized it in a way that may be useful to other medical institutes who are willing to pay for such information. This was also interested as these companies were almost able to make regular medical statistics and pieces of information into a lucrative business that other businesses have probably become highly dependent on.
-
http://online.wsj.com/public/resources/documents/info-presapp0605-31.html
I am an avid follower of political events. The current presidential race is no exception. Therefore I find that this data about previous president’s approval rate so fascinating. The data was put into the form of graphs making it easy to understand. This makes look for certain trends a much simpler task. This data is useful because many hypothesis can be drawn about the rates for the next president’s term. -
http://www.wisegeek.com/what-is-accounting-data.htm
Accounting data is a very important resource in today’s society. I was very interested in this article because it explained how accounting data is used to detect and prevent fraud and embezzlement. This is particularly interesting to me because I am an accounting major looking to get into the forensic field of accounting that specializes in crimes pertaining to accounting.
-
http://fivethirtyeight.com/features/a-day-in-the-life-of-an-iowa-family-drowning-in-campaign-ads/
The ad that I found on fivethirtyeight.com that I thought was interesting was a piece on a day in the life of an Iowa family, who like many others are bombarded with boatloads of ad campaigns in various forms. As a marketer, it is interesting to see how many ads this family had seen before even leaving the house; 15. Along with this concept, it is interesting to see what forms of advertising each candidate is using most. Trump seemed to send actual mail the most, Jeb Bush had the most billboards and Bernie Sanders dominated social media with every other post being about him. Another thing candidates don’t think of is how the family is set up in the house. This family had split political views, and one of the advertisements could end up being volatile in such an environment. By the end of the day, the family had seen around 90 ads collectively, which is something close to 5 ads per walking hour. In a world dominated by technology and social media, this number is only going to go up in years to come. The question for marketers is how to make their ads stick out from what is now considered “white noise” to most people. -
http://fivethirtyeight.com/features/lebrons-3-point-shot-has-abandoned-him/
Lebron James is one of the NBA’s most dominating players. With the league changing and players shooting more three point shots than ever, it is key that he follows the trend and dominates that part of the game. However, a recent article on fivethirtyeight.com challenges Lebron’s ability to do just that. The data shows that Lebron’s three point percentage has dropped in every aspect and every angle since last season. Needless to say, if Lebron wants a shot at a championship, he has to fix his shot first.
-
-
Shana Pote wrote a new post on the site MIS 0855: Data Science Spring 2016 8 years, 8 months ago
Here are the instructions in word (and as a PDF). Make sure you read them carefully!
When your assignment is complete, you’re going to email the document to me in .docx or .pdf […]
-
Shana Pote wrote a new post on the site MIS 0855: Data Science Spring 2016 8 years, 8 months ago
Here is the exercise.
And here is the spreadsheet you’ll need [In-Class Exercise 2.1 – 2015 Car Fuel Econ [Start]]
And here is the completed data dictionary for your reference In-Class Exercise 2.1 – 2015 C […]
- Load More