andrew_gelman_stats andrew_gelman_stats-2010 andrew_gelman_stats-2010-210 knowledge-graph by maker-knowledge-mining
Source: html
Introduction: Some recent blog comments made me rethink how to express disagreement when I write. I’ll first display my original blog entry, then some of the comments, then my reaction, and finally my new approach. As usual with negative comments, my first reaction was to be upset, but once the hurt feelings went away, I realized that these dudes had a point. Act 1 A few days I ago, I posted the following on 538: Self-described “ political operative ” Les Francis, with real-world experience as former executive director of the Democratic National Committee: I don’t need any polls to tell me that Republicans will do well in November. The “out” party almost always shows significant gains in the first midterm election of a new President. Political scientists Joe Bafumi, Bob Erikson, and Chris Wlezien, from elite, out-of-touch, ivory-tower institutions Dartmouth, Columbia, and Temple Universities: Game over. Act 2 After posting, I typically check the comments because so
sentIndex sentText sentNum sentScore
1 Act 1 A few days I ago, I posted the following on 538: Self-described “ political operative ” Les Francis, with real-world experience as former executive director of the Democratic National Committee: I don’t need any polls to tell me that Republicans will do well in November. [sent-4, score-0.797]
2 The “out” party almost always shows significant gains in the first midterm election of a new President. [sent-5, score-0.714]
3 Act 3 OK, I realized I needed more explanation, so I added a few paragraphs (see here ), saying what was on the horizontal and vertical axis of the graphs and explicitly stating that those graphs demonstrated that the polls are indeed informative in predicting the election. [sent-21, score-0.825]
4 In particular, many people thought that the main message of the graphs was that polls are more predictive of the vote as the election approaches. [sent-23, score-0.771]
5 Our point was that, even a year ahead of time, the generic ballot polls are good predictors, if appropriately piped through a linear model (as estimated from past data). [sent-27, score-0.817]
6 Here’s a try: Les Francis, who describes himself as a political operative real-world experience as former executive director of the Democratic National Committee, wrote this today on Frum Forum: I don’t need any polls to tell me that Republicans will do well in November. [sent-59, score-0.797]
7 The “out” party almost always shows significant gains in the first midterm election of a new President. [sent-60, score-0.714]
8 As a frequent user of polls myself, this jumped out at me. [sent-61, score-0.378]
9 It may be true that Francis doesn’t need any polls to draw his conclusions–although it would be to know this, as Francis is certainly aware of the polls at this time. [sent-62, score-0.807]
10 On the other hand, I think the polls can tell us a lot more than Francis gives them credit for. [sent-63, score-0.453]
11 Not 2002, since Bush’s party actually gained seats in that midterm election. [sent-65, score-0.536]
12 Perhaps the polls actually are useful in forecasting whether the out-party’s midterm gains will be closer to the 54 seats of 1994, the 15 seats of 1982, or the 2-seat pickup of 1962. [sent-73, score-1.303]
13 But Bafumi, Erikson, and Wlezien actually found out that the generic ballot predicts the election very well, even many months before the election. [sent-77, score-0.513]
14 The key is that you can’t just take the generic ballot number straight; you need to adjust it, which Bafumi et al. [sent-78, score-0.487]
15 Once you adjust for this (which you can do, using past election results), the generic polls are impressive predictors of the vote on Election Day. [sent-81, score-0.919]
16 To work with generic polls today, we could use one of the later graphs on the above grid to make a projection. [sent-94, score-0.802]
17 But my real point here is not the specific prediction but just that the polls tell us a lot. [sent-95, score-0.504]
18 Through some statistical analysis we can use the polls to learn a lot more than that the out party will likely show “significant gains” which might be somewhere between 2 and 54 seats in the House. [sent-96, score-0.737]
19 , can be valuable but he’s doing himself no favors by dismissing the polls or by making statements such as “Barack Obama’s 2008 winning margin was somewhat out of synch with the political alignment of the country at the time,” which is unsupported by any data I’ve seen. [sent-98, score-0.475]
20 (See, for example, this discussion , from nearly two years ago, on how the Democrats’ gains in House voting in 2008 were in fact as large as their gains in the presidential vote. [sent-99, score-0.452]
wordName wordTfidf (topN-words)
[('polls', 0.378), ('generic', 0.262), ('bafumi', 0.23), ('seats', 0.229), ('gains', 0.226), ('francis', 0.209), ('wlezien', 0.18), ('erikson', 0.169), ('graphs', 0.162), ('republicans', 0.143), ('party', 0.13), ('ballot', 0.126), ('election', 0.125), ('democrats', 0.121), ('house', 0.121), ('midterm', 0.119), ('click', 0.106), ('vote', 0.106), ('act', 0.104), ('les', 0.099), ('political', 0.097), ('operative', 0.088), ('comments', 0.076), ('pickup', 0.075), ('tell', 0.075), ('realized', 0.066), ('democratic', 0.065), ('elections', 0.061), ('better', 0.061), ('win', 0.061), ('significant', 0.06), ('reaction', 0.059), ('gained', 0.058), ('readers', 0.057), ('needed', 0.057), ('commenters', 0.056), ('director', 0.056), ('committee', 0.055), ('first', 0.054), ('joe', 0.054), ('executive', 0.052), ('swing', 0.051), ('september', 0.051), ('point', 0.051), ('need', 0.051), ('aren', 0.05), ('congressional', 0.048), ('adjust', 0.048), ('went', 0.047), ('forecasting', 0.047)]
simIndex simValue blogId blogTitle
same-blog 1 0.99999994 210 andrew gelman stats-2010-08-16-What I learned from those tough 538 commenters
Introduction: Some recent blog comments made me rethink how to express disagreement when I write. I’ll first display my original blog entry, then some of the comments, then my reaction, and finally my new approach. As usual with negative comments, my first reaction was to be upset, but once the hurt feelings went away, I realized that these dudes had a point. Act 1 A few days I ago, I posted the following on 538: Self-described “ political operative ” Les Francis, with real-world experience as former executive director of the Democratic National Committee: I don’t need any polls to tell me that Republicans will do well in November. The “out” party almost always shows significant gains in the first midterm election of a new President. Political scientists Joe Bafumi, Bob Erikson, and Chris Wlezien, from elite, out-of-touch, ivory-tower institutions Dartmouth, Columbia, and Temple Universities: Game over. Act 2 After posting, I typically check the comments because so
2 0.39064217 270 andrew gelman stats-2010-09-12-Comparison of forecasts for the 2010 congressional elections
Introduction: Yesterday at the sister blog , Nate Silver forecast that the Republicans have a two-thirds chance of regaining the House of Representatives in the upcoming election, with an expected gain of 45 House seats. Last month, Bafumi, Erikson, and Wlezien released their forecast that gives the Republicans an 80% chance of takeover and an expected gain of 50 seats. As all the above writers emphasize, these forecasts are full of uncertainty, so I treat the two predictions–a 45-seat swing or a 50-seat swing–as essentially identical at the national level. And, as regular readers know, as far back as a year ago , the generic Congressional ballot (those questions of the form, “Which party do you plan to vote for in November?”) was also pointing to big Republican gains. As Bafumi et al. point out, early generic polls are strongly predictive of the election outcome, but they need to be interpreted carefully. The polls move in a generally predictable manner during the year leading up to an
3 0.35425332 237 andrew gelman stats-2010-08-27-Bafumi-Erikson-Wlezien predict a 50-seat loss for Democrats in November
Introduction: They write : How many House seats will the Republicans gain in 2010? . . . Our methodology replicates that for our ultimately successful forecast of the 2006 midterm. Two weeks before Election Day in 2006, we posted a prediction that the Democrats would gain 32 seats and recapture the House majority. The Democrats gained 30 seats in 2006. Our current forecast for 2010 shows that the Republicans are likely to regain the House majority. . . . the most likely scenario is a Republican majority in the neighborhood of 229 seats versus 206 for the Democrats for a 50-seat loss for the Democrats . How do they do it? First, they predict the national two-party vote using the generic polls (asking voters which party they plan to vote for in the November congressional elections). Then they apply the national vote swing on a district-by-district level to predict the outcome in each district. They account for uncertainty in their predictions (I assume by using a model similar to what Gar
4 0.2891672 1512 andrew gelman stats-2012-09-27-A Non-random Walk Down Campaign Street
Introduction: Political campaigns are commonly understood as random walks, during which, at any point in time, the level of support for any party or candidate is equally likely to go up or down. Each shift in the polls is then interpreted as the result of some combination of news and campaign strategies. A completely different story of campaigns is the mean reversion model in which the elections are determined by fundamental factors of the economy and partisanship; the role of the campaign is to give voters a chance to reach their predetermined positions. The popularity of the random walk model for polls may be partially explained via analogy to the widespread idea that stock prices reflect all available information, as popularized in Burton Malkiel’s book, A Random Walk Down Wall Street. Once the idea has sunk in that short-term changes in the stock market are inherently unpredictable, it is natural for journalists to think the same of polls. For example, political analyst Nate Silver wrote
5 0.27634525 286 andrew gelman stats-2010-09-20-Are the Democrats avoiding a national campaign?
Introduction: Bob Erikson, one of my colleagues at Columbia who knows much more about American politics than I do, sent in the following screed. I’ll post Bob’s note, followed by my comments. Bob writes: Monday morning many of us were startled by the following headline: White House strenuously denies NYT report that it is considering getting aggressive about winning the midterm elections. At first I [Bob] thought I was reading the Onion, but no, it was a sarcastic comment on the blog Talking Points Memo. But the gist of the headline appears to be correct. Indeed, the New York Times reported that White House advisers denied that a national ad campaign was being planned. ‘There’s been no discussion of such a thing at the White House’ What do we make of this? Is there some hidden downside to actually running a national campaign? Of course, money spent nationally is not spent on targeted local campaigns. But that is always the case. What explains the Democrats’ trepidation abou
6 0.26756376 292 andrew gelman stats-2010-09-23-Doug Hibbs on the fundamentals in 2010
7 0.26459467 1570 andrew gelman stats-2012-11-08-Poll aggregation and election forecasting
8 0.24510908 364 andrew gelman stats-2010-10-22-Politics is not a random walk: Momentum and mean reversion in polling
9 0.23583846 394 andrew gelman stats-2010-11-05-2010: What happened?
10 0.23025738 2255 andrew gelman stats-2014-03-19-How Americans vote
11 0.1740448 384 andrew gelman stats-2010-10-31-Two stories about the election that I don’t believe
12 0.17046122 2343 andrew gelman stats-2014-05-22-Big Data needs Big Model
13 0.16440411 389 andrew gelman stats-2010-11-01-Why it can be rational to vote
14 0.16440411 1565 andrew gelman stats-2012-11-06-Why it can be rational to vote
15 0.15970799 1577 andrew gelman stats-2012-11-14-Richer people continue to vote Republican
17 0.149289 131 andrew gelman stats-2010-07-07-A note to John
19 0.14272273 551 andrew gelman stats-2011-02-02-Obama and Reagan, sitting in a tree, etc.
20 0.14180805 312 andrew gelman stats-2010-10-02-“Regression to the mean” is fine. But what’s the “mean”?
topicId topicWeight
[(0, 0.251), (1, -0.127), (2, 0.174), (3, 0.182), (4, -0.057), (5, -0.059), (6, -0.116), (7, -0.071), (8, -0.02), (9, -0.064), (10, 0.138), (11, 0.075), (12, 0.052), (13, -0.124), (14, -0.07), (15, -0.001), (16, -0.047), (17, -0.008), (18, -0.022), (19, 0.078), (20, -0.05), (21, 0.089), (22, -0.0), (23, 0.025), (24, -0.009), (25, 0.028), (26, 0.008), (27, 0.004), (28, -0.016), (29, 0.053), (30, -0.022), (31, 0.058), (32, -0.053), (33, -0.015), (34, 0.027), (35, 0.03), (36, -0.009), (37, 0.01), (38, 0.043), (39, -0.021), (40, -0.067), (41, 0.023), (42, 0.04), (43, 0.065), (44, -0.017), (45, -0.01), (46, -0.028), (47, -0.048), (48, 0.062), (49, 0.031)]
simIndex simValue blogId blogTitle
same-blog 1 0.95287573 210 andrew gelman stats-2010-08-16-What I learned from those tough 538 commenters
Introduction: Some recent blog comments made me rethink how to express disagreement when I write. I’ll first display my original blog entry, then some of the comments, then my reaction, and finally my new approach. As usual with negative comments, my first reaction was to be upset, but once the hurt feelings went away, I realized that these dudes had a point. Act 1 A few days I ago, I posted the following on 538: Self-described “ political operative ” Les Francis, with real-world experience as former executive director of the Democratic National Committee: I don’t need any polls to tell me that Republicans will do well in November. The “out” party almost always shows significant gains in the first midterm election of a new President. Political scientists Joe Bafumi, Bob Erikson, and Chris Wlezien, from elite, out-of-touch, ivory-tower institutions Dartmouth, Columbia, and Temple Universities: Game over. Act 2 After posting, I typically check the comments because so
2 0.90032953 292 andrew gelman stats-2010-09-23-Doug Hibbs on the fundamentals in 2010
Introduction: Hibbs, one of the original economy-and-elections guys, writes : The number of House seats won by the presidents party at midterm elections is well explained by three pre-determined or exogenous variables: (1) the number of House seats won by the in-party at the previous on-year election, (2) the vote margin of the in-partys candidate at the previous presidential election, and (3) the average growth rate of per capita real disposable personal income during the congressional term. Given the partisan division of House seats following the 2008 on-year election, President Obamas margin of victory in 2008, and the weak growth of per capita real income during the rst 6 quarters of the 111th Congress, the Democrats chances of holding on to a House majority by winning at least 218 seats at the 2010 midterm election will depend on real income growth in the 3rd quarter of 2010. The data available at this writing indicate the that Democrats will win 211 seats, a loss of 45 from the 2008 o
3 0.88896716 237 andrew gelman stats-2010-08-27-Bafumi-Erikson-Wlezien predict a 50-seat loss for Democrats in November
Introduction: They write : How many House seats will the Republicans gain in 2010? . . . Our methodology replicates that for our ultimately successful forecast of the 2006 midterm. Two weeks before Election Day in 2006, we posted a prediction that the Democrats would gain 32 seats and recapture the House majority. The Democrats gained 30 seats in 2006. Our current forecast for 2010 shows that the Republicans are likely to regain the House majority. . . . the most likely scenario is a Republican majority in the neighborhood of 229 seats versus 206 for the Democrats for a 50-seat loss for the Democrats . How do they do it? First, they predict the national two-party vote using the generic polls (asking voters which party they plan to vote for in the November congressional elections). Then they apply the national vote swing on a district-by-district level to predict the outcome in each district. They account for uncertainty in their predictions (I assume by using a model similar to what Gar
4 0.87393612 1512 andrew gelman stats-2012-09-27-A Non-random Walk Down Campaign Street
Introduction: Political campaigns are commonly understood as random walks, during which, at any point in time, the level of support for any party or candidate is equally likely to go up or down. Each shift in the polls is then interpreted as the result of some combination of news and campaign strategies. A completely different story of campaigns is the mean reversion model in which the elections are determined by fundamental factors of the economy and partisanship; the role of the campaign is to give voters a chance to reach their predetermined positions. The popularity of the random walk model for polls may be partially explained via analogy to the widespread idea that stock prices reflect all available information, as popularized in Burton Malkiel’s book, A Random Walk Down Wall Street. Once the idea has sunk in that short-term changes in the stock market are inherently unpredictable, it is natural for journalists to think the same of polls. For example, political analyst Nate Silver wrote
5 0.85067236 270 andrew gelman stats-2010-09-12-Comparison of forecasts for the 2010 congressional elections
Introduction: Yesterday at the sister blog , Nate Silver forecast that the Republicans have a two-thirds chance of regaining the House of Representatives in the upcoming election, with an expected gain of 45 House seats. Last month, Bafumi, Erikson, and Wlezien released their forecast that gives the Republicans an 80% chance of takeover and an expected gain of 50 seats. As all the above writers emphasize, these forecasts are full of uncertainty, so I treat the two predictions–a 45-seat swing or a 50-seat swing–as essentially identical at the national level. And, as regular readers know, as far back as a year ago , the generic Congressional ballot (those questions of the form, “Which party do you plan to vote for in November?”) was also pointing to big Republican gains. As Bafumi et al. point out, early generic polls are strongly predictive of the election outcome, but they need to be interpreted carefully. The polls move in a generally predictable manner during the year leading up to an
8 0.81112534 1000 andrew gelman stats-2011-11-10-Forecasting 2012: How much does ideology matter?
9 0.81092471 1570 andrew gelman stats-2012-11-08-Poll aggregation and election forecasting
10 0.80335891 43 andrew gelman stats-2010-05-19-What do Tuesday’s elections tell us about November?
12 0.78553069 364 andrew gelman stats-2010-10-22-Politics is not a random walk: Momentum and mean reversion in polling
13 0.78268552 286 andrew gelman stats-2010-09-20-Are the Democrats avoiding a national campaign?
14 0.77807081 312 andrew gelman stats-2010-10-02-“Regression to the mean” is fine. But what’s the “mean”?
15 0.77611864 934 andrew gelman stats-2011-09-30-Nooooooooooooooooooo!
17 0.75317472 1574 andrew gelman stats-2012-11-12-How to Lie With Statistics example number 12,498,122
18 0.7412886 1027 andrew gelman stats-2011-11-25-Note to student journalists: Google is your friend
19 0.73938543 300 andrew gelman stats-2010-09-28-A calibrated Cook gives Dems the edge in Nov, sez Sandy
20 0.73394603 1407 andrew gelman stats-2012-07-06-Statistical inference and the secret ballot
topicId topicWeight
[(5, 0.045), (9, 0.025), (15, 0.024), (16, 0.062), (21, 0.056), (24, 0.133), (34, 0.06), (45, 0.023), (50, 0.07), (76, 0.014), (77, 0.013), (86, 0.038), (93, 0.01), (95, 0.023), (99, 0.275)]
simIndex simValue blogId blogTitle
same-blog 1 0.9685573 210 andrew gelman stats-2010-08-16-What I learned from those tough 538 commenters
Introduction: Some recent blog comments made me rethink how to express disagreement when I write. I’ll first display my original blog entry, then some of the comments, then my reaction, and finally my new approach. As usual with negative comments, my first reaction was to be upset, but once the hurt feelings went away, I realized that these dudes had a point. Act 1 A few days I ago, I posted the following on 538: Self-described “ political operative ” Les Francis, with real-world experience as former executive director of the Democratic National Committee: I don’t need any polls to tell me that Republicans will do well in November. The “out” party almost always shows significant gains in the first midterm election of a new President. Political scientists Joe Bafumi, Bob Erikson, and Chris Wlezien, from elite, out-of-touch, ivory-tower institutions Dartmouth, Columbia, and Temple Universities: Game over. Act 2 After posting, I typically check the comments because so
Introduction: Jeff Ratto points me to this news article by Dean Baker reporting the work of three economists, Thomas Herndon, Michael Ash, and Robert Pollin, who found errors in a much-cited article by Carmen Reinhart and Kenneth Rogoff analyzing historical statistics of economic growth and public debt. Mike Konczal provides a clear summary; that’s where I got the above image. Errors in data processing and data analysis It turns out that Reinhart and Rogoff flubbed it. Herndon et al. write of “spreadsheet errors, omission of available data, weighting, and transcription.” The spreadsheet errors are the most embarrassing, but the other choices in data analysis seem pretty bad too. It can be tough to work with small datasets, so I have sympathy for Reinhart and Rogoff, but it does look like they were jumping to conclusions in their paper. Perhaps the urgency of the topic moved them to publish as fast as possible rather than carefully considering the impact of their data-analytic choi
3 0.94780892 1636 andrew gelman stats-2012-12-23-Peter Bartlett on model complexity and sample size
Introduction: Zach Shahn saw this and writes: I just heard a talk by Peter Bartlett about model selection in “unlimited” data situations that essentially addresses this curve. He talks about the problem of model selection given a computational budget (rather than given a sample size). You can either use your computational budget to get more data or fit a more complex model. He shows that you can get oracle inequalities for model selection algorithms under this paradigm (as long as the candidate models are nested). I can’t follow all the details but it looks cool! This is what they should be teaching in theoretical statistics class, instead of sufficient statistics and the Neyman-Pearson lemma and all that other old stuff. Zach also asks: I have a question about political science. I always hear that the direction of the economy is one of the best predictors of election outcome. What’s your thinking about the causal mechanism(s) behind the success of economic trend indicators as pr
4 0.9475711 1793 andrew gelman stats-2013-04-08-The Supreme Court meets the fallacy of the one-sided bet
Introduction: Doug Hartmann writes ( link from Jay Livingston): Justice Antonin Scalia’s comment in the Supreme Court hearings on the U.S. law defining marriage that “there’s considerable disagreement among sociologists as to what the consequences of raising a child in a single-sex family, whether that is harmful to the child or not.” Hartman argues that Scalia is factually incorrect—there is not actually “considerable disagreement among sociologists” on this issue—and quotes a recent report from the American Sociological Association to this effect. Assuming there’s no other considerable group of sociologists (Hartman knows of only one small group) arguing otherwise, it seems that Hartman has a point. Scalia would’ve been better off omitting the phrase “among sociologists”—then he’d have been on safe ground, because you can always find somebody to take a position on the issue. Jerry Falwell’s no longer around but there’s a lot more where he came from. Even among scientists, there’s
5 0.94612956 1723 andrew gelman stats-2013-02-15-Wacky priors can work well?
Introduction: Dave Judkins writes: I would love to see a blog entry on this article , Bayesian Model Selection in High-Dimensional Settings, by Valen Johnson and David Rossell. The simulation results are very encouraging although the choice of colors for some of the graphics is unfortunate. Unless I am colorblind in some way that I am unaware of, they have two thin charcoal lines that are indistinguishable. When Dave Judkins puts in a request, I’ll respond. Also, I’m always happy to see a new Val Johnson paper. Val and I are contemporaries—he and I got our PhD’s at around the same time, with both of us working on Bayesian image reconstruction, then in the early 1990s Val was part of the legendary group at Duke’s Institute of Statistics and Decision Sciences—a veritable ’27 Yankees featuring Mike West, Merlise Clyde, Michael Lavine, Dave Higdon, Peter Mueller, Val, and a bunch of others. I always thought it was too bad they all had to go their separate ways. Val also wrote two classic p
6 0.94592917 61 andrew gelman stats-2010-05-31-A data visualization manifesto
7 0.94416893 1981 andrew gelman stats-2013-08-14-The robust beauty of improper linear models in decision making
9 0.94302505 2140 andrew gelman stats-2013-12-19-Revised evidence for statistical standards
10 0.9424504 956 andrew gelman stats-2011-10-13-Hey, you! Don’t take that class!
11 0.94239634 2341 andrew gelman stats-2014-05-20-plus ça change, plus c’est la même chose
12 0.94166607 1162 andrew gelman stats-2012-02-11-Adding an error model to a deterministic model
13 0.94155931 2281 andrew gelman stats-2014-04-04-The Notorious N.H.S.T. presents: Mo P-values Mo Problems
14 0.94154966 1634 andrew gelman stats-2012-12-21-Two reviews of Nate Silver’s new book, from Kaiser Fung and Cathy O’Neil
16 0.94116843 670 andrew gelman stats-2011-04-20-Attractive but hard-to-read graph could be made much much better
17 0.94054556 292 andrew gelman stats-2010-09-23-Doug Hibbs on the fundamentals in 2010
19 0.93984807 2288 andrew gelman stats-2014-04-10-Small multiples of lineplots > maps (ok, not always, but yes in this case)
20 0.93917429 1734 andrew gelman stats-2013-02-23-Life in the C-suite: A graph that is both ugly and bad, and an unrelated story