Status
Available
Call number
Publication
Dey Street Books (2017), 352 pages
Description
A former Google data scientist presents an insider's look at what the vast, instantly available amounts of information from the Internet can reveal about human civilization and society.
User reviews
LibraryThing member little-gidding
This book is in that particular genre where the author tries to make his or her area of expertise (often physics for some reason, though clearly not in this case) palatable and accessible to the "common (wo)man." These types of books fail when the author doesn't dumb it down enough or dumbs it down
At the beginning of the book, my inner skeptic was anxiously asking about correlation vs causation and how people can know they're asking the right questions of the right data. By the end of the book, Stephens-Davidowitz had satisfactorily addressed most of my initial concerns and provided some insight into data science, social science, and some aspects of human nature along the way. Plus, the book made me laugh (well, chuckle) out loud more than a few times, which means I was pretty engaged and is not bad for a book about data science.
Some notes:
- The subtitle ("Big Data, new data, and what the internet can tell us about who we really are") is slightly misleading. While much of the book does rely on search queries (predominately Google) and Twitter and Facebook updates, plenty of the analysis and studies rely on non-internet data sources. Stephens-Davidowitz is clearly excited about all of the new ways to use all of the new internet data, but the overall focus of the book is on Big Data of all kinds and its powers and drawbacks.
- Some chapters illustrate the fact that people admit things on the internet they would not admit elsewhere. Issues addressed include porn preferences and racism, both discussed in detail, and child abuse, discussed in less detail. Although the possible conclusions range from unsavory to downright depressing, the topics are relevant to addressing the book's points about data and social science; however, worth noting because some readers will be sensitive to these topics.
(Thank you, Dey Street Books and GoodReads for the ARC.)
Show More
too much. Stephens-Davidowitz's area is economics/social science by way of Big Data, and he dumbs it down just the right amount.At the beginning of the book, my inner skeptic was anxiously asking about correlation vs causation and how people can know they're asking the right questions of the right data. By the end of the book, Stephens-Davidowitz had satisfactorily addressed most of my initial concerns and provided some insight into data science, social science, and some aspects of human nature along the way. Plus, the book made me laugh (well, chuckle) out loud more than a few times, which means I was pretty engaged and is not bad for a book about data science.
Some notes:
- The subtitle ("Big Data, new data, and what the internet can tell us about who we really are") is slightly misleading. While much of the book does rely on search queries (predominately Google) and Twitter and Facebook updates, plenty of the analysis and studies rely on non-internet data sources. Stephens-Davidowitz is clearly excited about all of the new ways to use all of the new internet data, but the overall focus of the book is on Big Data of all kinds and its powers and drawbacks.
- Some chapters illustrate the fact that people admit things on the internet they would not admit elsewhere. Issues addressed include porn preferences and racism, both discussed in detail, and child abuse, discussed in less detail. Although the possible conclusions range from unsavory to downright depressing, the topics are relevant to addressing the book's points about data and social science; however, worth noting because some readers will be sensitive to these topics.
(Thank you, Dey Street Books and GoodReads for the ARC.)
Show Less
LibraryThing member nmarun
It's hard to imagine a book written mostly based on what people entered in Google search, but the author ingeniously used this easy-to-obtain repository to find many things about us.
Obtaining an authentic source of data is key to any data analysis. He rightly argues that people are less truthful to
There's a disclaimer on what Big Data cannot do and how it should not be used. The author also warns us of jumping to conclusions just because our data set was large.
I thoroughly enjoyed reading this book and will wait for Everybody (still) lies.
Obtaining an authentic source of data is key to any data analysis. He rightly argues that people are less truthful to
Show More
their closest aides/partners, but disclose their honest feelings to a search engine. I'm guilty myself, so much so, that I'm cautious of what I enter in the search bar now.. well not quite, I'll probably continue being myself.There's a disclaimer on what Big Data cannot do and how it should not be used. The author also warns us of jumping to conclusions just because our data set was large.
I thoroughly enjoyed reading this book and will wait for Everybody (still) lies.
Show Less
LibraryThing member ohernaes
Entertaining and interesting. Maybe jumping to conclusions a little fast, e.g. Obama lost four percent of votes in areas with racist searches. He often takes google searches to be random samples of people's thoughts, which they are not, and often does not seem not to worry about selection in who
Falsifiable Freud: phallos shaped fruit like banana and cucumber not more common than other fruits and vegetables in dreams. But this does not really falsify, everything may mean something. Better on Freudian slips - typing errors with sexual connotations not unusual compared to random. But a Freudian will presumably still believe it does mean something when a human commits such an error... The book loses focus somewhat when starting to talk about big data generally. Good and sober section on the dangers and desirability/fairness of using big data information to assess loan applications etc. Recommended.
Show More
searches and why. Fun: seeming breastfeeding wife fetish in India, vagina smell most searched by women, English vs Spanish autocomplete with pregnant wife; but could discuss more what drives results. Falsifiable Freud: phallos shaped fruit like banana and cucumber not more common than other fruits and vegetables in dreams. But this does not really falsify, everything may mean something. Better on Freudian slips - typing errors with sexual connotations not unusual compared to random. But a Freudian will presumably still believe it does mean something when a human commits such an error... The book loses focus somewhat when starting to talk about big data generally. Good and sober section on the dangers and desirability/fairness of using big data information to assess loan applications etc. Recommended.
Show Less
LibraryThing member bemislibrary
The book does not so much tell us about how people really are as much as tell how the internet can be used to support personal beliefs and ideologies. The scenarios provided lessons on how to manipulate data. The author believes a proper conclusion sums up and set parameters for the next steps. He
I was randomly chosen through a Goodreads Giveaway to receive this book free from the publisher. Although encouraged, I was under no obligation to write a review. The opinions I have expressed are my own.
Show More
advises readers to follow what people do and not what they say. Throughout the author talks about what he has done or should do. The notes provide an interested look at how the he used data to support theorems but not why he selected specific propositions. It was interesting to read, but as the title says, it could all be a lie. I was randomly chosen through a Goodreads Giveaway to receive this book free from the publisher. Although encouraged, I was under no obligation to write a review. The opinions I have expressed are my own.
Show Less
LibraryThing member jpsnow
Everybody lies, and a lot of the conventional wisdom cast at us is also a lie. This book breaks new ground itself. Moreover, it shows just how much new ground is being broken in the field of big data analytics. As an experienced data analyst, what I most appreciated is Stephens-Davidowitz's ability
Show More
to show how it's not so much the bigness of modern data that matters, but the new sources and our improved ways of approaching problems. From Google searches to sentiment analysis, we can tell things about ourselves that weren't possible even ten years ago. The emerging picture is proving out that we fool others and even ourselves quite often. Show Less
LibraryThing member wagner.sarah35
This book provides plenty of food for thought and ambitious in its scope - the author uses Google search data to present theories about why Donald Trump was elected president, the prevalence of racism in American society, the value of attending an elite high school, how to determine a good sport
Show More
player, and more topics both grand and petty. It's all presented in an engaging manner and helps one make sense of what some of the trends identified mean. It also can sometimes be a little creepy if one thinks about how often we interact with internet services (Google, Facebook, Twitter, etc) and the amount of data that is recorded about our behavior (even if it's largely anonymous data gathering). Show Less
LibraryThing member Tytania
Jam-packed with "who'd have thought it?" insights based on his professional data analysis skills, and reams of data, mostly Google searches. A wowser on nearly every page, many which you can't resist sharing. What's the magic age for a person to be for his team's World Series win to make him a
Show More
lifelong fan? How could you have predicted where Trump would win based on offensive Google searches? Like Freakonomics meets Malcolm Gladwell. Fun! Show Less
LibraryThing member tjsjohanna
The biggest appeal of books like these for me is finding out all the things that I thought I knew were false. The author shares a great sample of interesting things that illustrate the ways really big data groups can teach us things about ourselves that we didn't know or didn't want to know. This
Show More
work did make me wonder if I should get off the internet and stay off! Show Less
LibraryThing member asxz
Hard to tell if the author does good science because he does such uninteresting writing. I can imagine another book that could be so pleased with itself and, at the same time, so calculatingly composed to fulfill a contractual obligation. There is very little that is remarkable here, except for
This was weak, sub-Gladwell, smug pop science. Not recommended.
Show More
perhaps the author's obsession with porn. But even those endlessly prurient sections felt like someone trying to get more clicks for a lightweight article rather than actually contributing to the overarching theory of a book.This was weak, sub-Gladwell, smug pop science. Not recommended.
Show Less
LibraryThing member Narilka
In Everybody Lies, Seth Stephens-Davidowitz explores the idea behind social desirability bias and how internet searches are helping Big Data paint a clearer picture about society. In short:
Many people under-report embarrassing behaviors and thoughts on surveys. They want to look good, even though
Stephens-Davidowitz posits that while people may lie to anonymous surveys they tend to type their true feelings and intentions into Google searches. It is this vast sum of new data that will allow researchers to make better predictions and offers brand new tools to allow insight all aspects of human behavior that direct questioning never could. It's a fascinating idea and the book provides plenty of food for thought.
The new age of Big Data is starting to show how wrong many of our assumptions about society are. How Google searches predicted Donald Trump's victory to common body anxieties to why people root for specific sports teams to the value of attending an elite high school to zooming in on health data and how it could change the way we receive care. It's eerie and a bit creepy when you stop and think about what people type into an internet search box, how much of that data is being captured and just what that data is starting to say about society. On the flip side, the author notes that Big Data has many pitfalls and it's a fairly new science that is still in its infancy.
Many people under-report embarrassing behaviors and thoughts on surveys. They want to look good, even though
Show More
most surveys are anonymous.Stephens-Davidowitz posits that while people may lie to anonymous surveys they tend to type their true feelings and intentions into Google searches. It is this vast sum of new data that will allow researchers to make better predictions and offers brand new tools to allow insight all aspects of human behavior that direct questioning never could. It's a fascinating idea and the book provides plenty of food for thought.
The new age of Big Data is starting to show how wrong many of our assumptions about society are. How Google searches predicted Donald Trump's victory to common body anxieties to why people root for specific sports teams to the value of attending an elite high school to zooming in on health data and how it could change the way we receive care. It's eerie and a bit creepy when you stop and think about what people type into an internet search box, how much of that data is being captured and just what that data is starting to say about society. On the flip side, the author notes that Big Data has many pitfalls and it's a fairly new science that is still in its infancy.
Show Less
LibraryThing member addunn3
The author looks at ways big data can be used.
LibraryThing member waldhaus1
Part of what for my interest was the answers the author got for some of life’s persistent questions. He concludes by suggesting big data will place the soft sciences on more solid footing.
I found myself wondering how he got access to the data and how google keeps track of so many questions I
I found myself wondering how he got access to the data and how google keeps track of so many questions I
Show More
think of as unusual. I guess that odd part of Google’s business. Show Less
LibraryThing member nicdevera
A little disjointed, some of the humor falls flat. Worth reading for the tidbits, some glimpses into Google and Facebook's inner workings. Big Data is the future, obviously.
LibraryThing member LisCarey
An ongoing problem in research in psychology, political polling, and many other areas that rely on asking people questions about their views, activities, and experiences, is that people lie. Sometimes because the topic is a sensitive one, sometimes because they don't like pollsters, sometimes
This audiobook is about what you can find when you look at the the sources of data where people don't lie, because it would defeat their purpose rather than yours.
Google and other search engines are major sources of that "honest data." When people are looking for information, whether they're looking for help with their depression, or for racist jokes, they won't get what they want if they don't say what they want clearly enough for Google or Bing or DuckDuckGo to find it. Even on less charged or sensitive subjects, though, the search engines get far more data on any given topic than any manageable survey could retrieve in a workable timeframe.
And that data can tell us important things, trivial things, and things we might prefer not to know, such as where those searches for racist jokes are coming from, and what they tell us about who voted against Barack Obama, and why. Or where specific public needs aren't being met. It can tell us what are the symptoms people have when they are in the very early stages of pancreatic cancer, early enough that it might still be beatable.
To be clear, Stephens-Davidowitz looks at other sources of "big data" too, not just the search engines by any means, He also looks seriously at both the good and the bad that intelligent use of big data can offer; we want the government to be able to respond to public needs, for instance, but we don't want the Minority Report vision of a future with a Bureau of Pre-Crime, arresting you before you've even decided to do anything criminal.
Mostly it's lively and entertaining, as well as thought-provoking. Unfortunately for me, I was listening to it on Election Day and the couple of days following, when we didn't know what the outcome would be, and the section talking about the 2016 election was very hard on my nerves, and I nearly stopped listening. But that's my sensitivities and the timing, and I expect that section won't have that effect on most listeners.
I did find the conclusion to be unintentionally funny, as he went on and on about the importance of a properly strong conclusion not being verbose... But I forgive him for that. Recommended.
I bought this audiobook.
Show More
simply as a joke. Whatever the reason, a significant percentage of the people responding to any survey, will lie, and undermine the value of the data you think you're gathering.This audiobook is about what you can find when you look at the the sources of data where people don't lie, because it would defeat their purpose rather than yours.
Google and other search engines are major sources of that "honest data." When people are looking for information, whether they're looking for help with their depression, or for racist jokes, they won't get what they want if they don't say what they want clearly enough for Google or Bing or DuckDuckGo to find it. Even on less charged or sensitive subjects, though, the search engines get far more data on any given topic than any manageable survey could retrieve in a workable timeframe.
And that data can tell us important things, trivial things, and things we might prefer not to know, such as where those searches for racist jokes are coming from, and what they tell us about who voted against Barack Obama, and why. Or where specific public needs aren't being met. It can tell us what are the symptoms people have when they are in the very early stages of pancreatic cancer, early enough that it might still be beatable.
To be clear, Stephens-Davidowitz looks at other sources of "big data" too, not just the search engines by any means, He also looks seriously at both the good and the bad that intelligent use of big data can offer; we want the government to be able to respond to public needs, for instance, but we don't want the Minority Report vision of a future with a Bureau of Pre-Crime, arresting you before you've even decided to do anything criminal.
Mostly it's lively and entertaining, as well as thought-provoking. Unfortunately for me, I was listening to it on Election Day and the couple of days following, when we didn't know what the outcome would be, and the section talking about the 2016 election was very hard on my nerves, and I nearly stopped listening. But that's my sensitivities and the timing, and I expect that section won't have that effect on most listeners.
I did find the conclusion to be unintentionally funny, as he went on and on about the importance of a properly strong conclusion not being verbose... But I forgive him for that. Recommended.
I bought this audiobook.
Show Less
LibraryThing member Paul_S
Jumping to conclusions: the book. Chapters could be sold as clickbait articles.
LibraryThing member amberwitch
Interesting book, but not that engaging.
LibraryThing member AKBouterse
An interesting book to have read after taking a social statistics class this last semester. I don’t agree with all his conclusions about the future of research and I would like to see additional evidence for some of his research but overall an interesting book.
LibraryThing member lpg3d
While this book is a somewhat sort and easy read, it is nevertheless an important read. Big data is everywhere now, and many of the details of our lives are now recorded in databases. [b:Everybody Lies: Big Data, New Data, and What the Internet Can Tell Us About Who We Really Are|28512671|Everybody
(Note: This book was provided to me by the publisher via a Goodreads giveaway.)
Show More
Lies Big Data, New Data, and What the Internet Can Tell Us About Who We Really Are|Seth Stephens-Davidowitz|https://images.gr-assets.com/books/1489069766s/28512671.jpg|48667223] explores the many uses of this data to learn what's really going on (or not) in our homes and bedrooms. It illuminates the extent that everyone lies to each other and themselves, by showing that our responses to surveys and other voluntary studies are often false. Some of these results are depressing, but all are enlightening. The book also explores the way that big data may be abused against us, but provides little hope for any way of preventing this abuse. A good, important, and timely book.(Note: This book was provided to me by the publisher via a Goodreads giveaway.)
Show Less
LibraryThing member ms_rowse
Fascinating and also a little horrifying. Important read to understand how data is used to influence behaviors. I appreciated that Stephen-Davidowitz acknowledged the myriad ethical implications to consider when using and collecting data. If anything, this book reaffirmed my position that STEM
Show More
careers cannot exist in a vacuum--we need the humanities alongside STEM to remind us that while data might help us make sense of our world, we aren't robots. Nuance and ethics are still important to our survival as a species. Show Less
LibraryThing member Kavinay
The upshot of this book is not that big data is the holy grail. Rather, the recurring theme in all of Stephens-Davidowitz's interesting examples is just that most self-reporting is awful.
I'm still skeptical about the big data revolution--and this book doesn't really focus on implicit bias in
I'm still skeptical about the big data revolution--and this book doesn't really focus on implicit bias in
Show More
analysis of large data sets--but the conventional research methods of social sciences are amusingly torn to pieces (much like advertising ROI was absolutely shredded in the digital age where measurement was no longer entirely by gut). Show Less
LibraryThing member wvlibrarydude
A light read regarding big data. He covers the positive uses and pitfalls of analyzing big data. The research does raise questions about the subjects covered. I just wish more methodology in the research used was shared.
Awards
Chicago Public Library Best of the Best: Adults (Selection — 2017)
Language
Pages
352
Original language
English
Original publication date
2017-05-09
ISBN
0062390856 / 9780062390851