とある学生がノートを公開しているようです。
https://docs.google.com/file/d/0B4JUNQtZHukBeWdOVTR5NjEzM2s/edit
いつまで見れるのかはわからないので興味があればどうぞお早めにDLしてください。
(April 7th)：時間選好・
2014年4月8日 7:26
この手の話は、
このメモを公開した理由は、
個人的には、
そのメカニズムを明らかにして介入方法を開発できれば、
ちなみにJSHINE（http://park.itc.u-
誰か解析進めているんでしょうか？
とても一人でできる気がしないので、
さて、
時間選好については日本では阪大の大竹先生たちのグループが詳し
本も出ているので興味ある人はアマゾンでどうぞ。
アウトカムは健康ではなく教育の何かですが、
この手の話は、数式や数字や条件節がたくさん出てきます。
個人的にはこのレクチャーを英語でフォローしていくのは極めて困
元々内容を知っているのでなんとなく理解できますが、
時間選好は、
実際授業中も、日本人ではない学生から（
どんだけ近視的なのかそれとも将来が見えないのかそれとも明日の
個人的には、来年せいぜい110ドルくらいでも、
時間選好の国際比較のデータはすでに論文であったかと思いますが
時代もかわったら同一国内でも利用できないかも。
今の日本で調査をしたら、かなり時間選好が低く出るでしょう（
それは日本人そのものが保守的な考えであることだけではなく、
さて、時間選好は、マイナス６％
個人的にもマイナスはずっと疑問だった＆
外科レジデントの例で、「
まぁあと外科のレジデントに限らず、
子どもがいて今年お金が必要な人は、
つまり、同じ人でもタイミングにより異なりそうです。
しかし個人的にお金が必要になったからと言って、
つまり、
まぁそういう誤差も集団で見れば相殺するという考えもあるでしょ
長文につかれてきたので一気に割愛しますが（苦笑）、
などいろいろ考えた2時間でした。
Explaining Health Behavior: Insights from Behavioral EconomicsのTopic: Can we nudge our way to better food choiceのメモ
Guest Lecturer: Christina Roberto, Assistant Professor, SBS.
Ecuador becomes the first country in Latin America to approve traffic light labelling
上記のエクアドルの記事ですが、
"Ecuador traffic light labelling"でグーグルで画像検索すると例が見れます。
例）
しかしながら、上記の記事には言及ありませんが、授業によると、
食品のラベルを人は実際どれくらい見ているのか？については、
別の某調査によると、「食品ラベルをみている」
まぁ結局ラベルだけだとだめじゃね？
Evaluating the impact of menu labeling on food choices and intake.
Roberto CA1, Larsen PD, Agnew H, Baik J, Brownell KD.
PMID: 20019307
Am J Public Health. 2010 Feb;100(2):312-8. doi: 10.2105/AJPH.2009.160226. Epub 2009 Dec 17.
ちょっと専門的な話になりますが、
(1) a menu without any calorie labels (no calorie labels);何もしない
(2) a menu with calorie labels (calorie labels); メニューに●●カロリーと表示
(3)
a menu with calorie labels and a label at the top left corner of the
menu that read, ‘‘The recommended daily caloric intake for an average
adult is 2000 calories’’ (calorie labels plus
information).メニューに●●カロリーと表示して、
んで、彼らに自由に夕食を食べてもらい、ラベリング、および、
論文を読んで面白いと思ったのは、この研究では、
ラベリング群では、帰宅後にデザートで”しっかり補充している”
彼女の主張は「覚えていない（覚えにくい）
まぁそうですよね、いくら「この商品には塩分５グラムです」
そもそも覚えにくいものなんて覚えれないよね、ということです。
確かに、医者でさえ、
そこで以前にも紹介したMy plateみたいなものに到達するのですね。
（この画像は以下のページ http://www.rivertea.com/blog/
日本の素晴らしく複雑怪奇な「食事バランスガイド」も、
他にも多数紹介あり、おもしろいね。
カナダ
中国
フランス
ドイツ
ハンガリーはシンプル
砂糖が体に悪いことをアピールする対照的な２つの動画。
The Real Bears（楽しい）https://www.youtube.
5 Extra Years（恐怖）https://www.youtube.
どっちが効果的か？と。まぁよくわかりませんが、
ちょっと前に話題になったNYCでの砂糖入りの飲み物のサイズ規
そしてそれに反対する（
「BAN(廃止)」や「for the choice（選択権のために）」とかなんだかアメリカ人が（
ちなみに一般の人には「research」
まぁ今日の話は、かなり国民差がありそうです。
日本をよくするには、日本で独自の研究が必要だと思います。
日本はアメリカと違い、教育レベルもかなり高く、
そして今日の本論とは関係ありませんが、
資料を配布しないと、
Readings:
1.
Chandon P, & Wansink B. Does food marketing need to make us fat? A
review and solutions. Nutrition Reviews, 2012;70: 571-593.
2.
Schwartz J, Riis J, Elbel B, Ariely D. Inviting consumers to downsize
fast-food portions significantly reduces calorie consumption. Health
Affairs, 2012;31:399-407.
3.
Liu P.J., Wisdom, J., Roberto, C.A., Liu, L.J., & Ubel, P.A. (in
press). Using behavioral economics to design more effective food
policies to address obesity. Applied Economic Perspectives and
Management.
追記：オバマ夫人がエルモなどと健康増進の運動に参加している図
The First Lady, Elmo, and Rosita Partner to Encourage Healthy F
SBS503BehavioralEconomicsSessi
過去の２つのノートとタイトルの標準化が図れていないのが問題で
今日も講義資料が公開されていないゲストスピーカーだったのでメ
いつも講義を聞いているときには、
1.30-2.20 pm
Topic: Behavioral approaches to reducing sugar-sweetened beverage consumption.
Guest Lecturer: Jason Block, Assistant Professor of Medicine, Harvard Medical School
なぜ子どもは砂糖入りの飲み物を飲むか？
おいしい、やすい、元気が出る、仲間と集まる、
Block
JP, Gillman MW, Linakis SK, Goldman RE. "If it tastes good, I'm
drinking it": qualitative study of beverage consumption among college
students. J Adolesc Health. 2013 Jun;52(6):702-6. Epub 2013 Feb 14.
Journal of Adolescent Health 52 (2013) 702e706
以下、めんどくさいので sugar-sweetened beverage (SSB) と省略。
次の介入によりSSBの消費をどのように減らせるか？
１）課税する ２）行動経済学 ３）教育
１）課税する
アメリカは、州によって、食べ物・水・砂糖を含んだ飲み物、
sales taxとlarge size taxの違いは？：後者は製造者に課す税金→
BWH（近所のHarvard の関連病院）のカフェテリアで実験：SSBに３５％の税金。
→値段をあげると、普通のソーダの売り上げが減って、
他の国でも同じようなことをやった研究結果。
・J Health Econ. 2013 Jan;32(1):219-39. doi: 10.1016/j.jhealeco.2012.10.
Implications of a sugar-sweetened beverage (SSB) tax when substitutions to non-beverage items are considered.
Finkelstein EA1, Zhen C, Bilger M, Nonnemaker J, Farooqui AM, Todd JE.
・BMJ. 2013 Oct 31;347:f6189. doi: 10.1136/bmj.f6189.
Overall
and income specific effect on prevalence of overweight and obesity of
20% sugar sweetened drink tax in UK: econometric and comparative risk
assessment modelling study.
Briggs AD1, Mytton OT, Kehlbacher A, Tiffin R, Rayner M, Scarborough P.
実際は同じ値段でも、税込み価格と税抜き価格、
SALIENCE AND TAXATION:
THEORY AND EVIDENCE
Raj Chetty
Adam Looney
Kory Kroft
Working Paper 13330
NATIONAL BUREAU OF ECONOMIC RESEARCH
２）行動経済学
人は飲み物にカロリーがあると想定しないこともあるらしい。
Int J Obes Relat Metab Disord. 2000 Jun;24(6):794-
Liquid versus solid carbohydrate: effects on food intake and body weight.
DiMeglio DP1, Mattes RD.
飲み物の配置を、
ゼロキロカロリーの飲み物は、SSBの課税により売上上昇。
→（坪谷の推測）配置換えは、何であっても飲む気をなくさせる？
配置により人の行動がどのように変わるか？
例：アイスクリームは隠すことで売り上げを減らせる
３）教育の効果
BWHのカフェテリアでやって実験。各食品に赤・緑・
結果：食べ物の購入の分布は変わらず。
ちなみに購入の波形はAT（心房頻拍）の心電図みたいだと（
他の教育方法についての動画（基本的には“恐怖系“。）
Readings:
- Block JP, Chandra A, McManus KD, Willett WC. Point-of-purchase price and education intervention to reduce consumption of sugary soft drinks. Am J Public Health. 2010 Aug;100(8):1427-33. doi: 10.2105/AJPH.2009.175687.
- Block
JP, Gillman MW, Linakis SK, Goldman RE. "If it tastes good, I'm
drinking it": qualitative study of beverage consumption among college
students. J Adolesc Health. 2013 Jun;52(6):702-6. doi: 10.1016/j.jadohealth.2012.11.
017.
- Block JP, Condon SK, Kleinman K, Mullen J, Linakis S, Rifas-Shiman S, Gillman MW. Consumers' estimation of calorie content at fast food restaurants: cross sectional observational study. BMJ. 2013 May 23;346:f2907. doi: 10.1136/bmj.f2907.
- Blcok JP. A substantial tax on sugar sweetened drinks could help reduce obesity. BMJ. 2013 Oct 31;347:f5947. doi: 10.1136/bmj.f5947.
その２：今日は前半と後半で違う人が講義
2.30-3.20 pm
Topic: Behavioral economic interventions to promote healthy food choices
Guest Lecturer: Anne Thorndike, Assistant Professor of Medicine, Massachusetts General Hospital
今日の話す内容
１）background: obesity, healthy food choices and BE
２）traffic light labels and choice architecture(配置を変えること)
３）social norms and financial incentives
４）BE in a community setting
１）
USAの州別
肥満は遺伝子では説明できない以上に増えていると（
彼女は、
HOST: non-modifiable:gene, age, gender/ modifiable:behaviors & attitude
Env: physical , economic sociocultural, policy
Vectors; PC, cars, energy-dense food, drinks
古典的経済学と行動経済学の違いの復習
Traditional econ（TE):
Behavioral econ:なわけない。
例
１）status quo：金曜日はピザ
２）今の本能に従ってしまう（
など
カロリーを表示させることは、BEか？
→いやちがうでしょう。
２）
*traffic light labels(TLL) and choice architecture(配置を変えること)
TLLはMGHでやった。詳細は上にも書いてあるリンク。
MGHは雇用２３０００人、カフェテリア利用者は６０００人。
Platinum plate には、年齢性別職種などの情報が入っているので、
Phase1）ラベリング
Phase2）１から３か月後。Chioce artchitetureをやった。
Am J Public Health. 2012 Mar;102(3):527-
A 2-phase labeling and choice
Thorndike AN1, Sonnenberg L, Riis J, Barraclough S, Levy DE.
ベースライン→P1でGが増えて、Rは減った。
人種別＆ジョブタイプ別にも評価。
効果が長期間持続するかも評価した。
Am J Prev Med. 2014 Feb;46(2):143-9. doi: 10.1016/j.amepre.2013.10.002.
Traffic-light labels and choice architecture: promoting healthy food choices.
Thorndike AN1, Riis J2, Sonnenberg LM3, Levy DE4.
Conclusions: A traffic-light and choice architecture cafeteria intervention resulted in sustained
healthier choices over 2 years, suggesting that food environment interventions can promote longterm
changes in population eating behaviors.(Am J Prev Med 2014;46(2):143–149)
３）
*social norms and financial incentives
参照論文
・Social norms and energy conservation_Hunt Allcott
・A
2-Phase Labeling and Choice Architecture Intervention to Improve
Healthy Food and Beverage Choices Anne N Thorndike_Am J Public
Health_2012
４）
*BE in a community setting
MGHは特別なセッティングなので、一般化できるかな？
MGHとChelsea community centorとの協力で行っている（
Project 1 Chelsea corner, 35000residents
62%Latin、健康的な食品を変えないがけっこうある。
６つの店、３（介入）と３にわける
アウトカム１）WICフルール野菜の消費、２）
裏話：各店に協力金として５００ドル支払いした。
Project ２ Choose well/Elige bien!
１）全ての商品へのラベリングが有効かどうかを見る
２）経済的なインセンティブが有効かを確認する
アウトカム
１）飲み物の赤の消費量
２）飲み物の赤／緑の比率
対象：２１６人のRCT、Loyalty card５％値引き
ベスライン２か月、介入（ラベリング、
９８％女性、９７＆ラテン、６８％が店を週に２回以上利用、
Traffic light rating system
上記にまつわるニュース
Readings
Thorndike AN, Riis J, Sonnenberg LM, Levy DE. Traffic-light labels and choice architecture: promoting healthy food choices. Am J Prev Med. 2014 Feb;46(2):143-9. doi: 10.1016/j.amepre.2013.10.002.
Sonnenberg
L, Gelsomin E, Levy DE, Riis J, Barraclough S, Thorndike AN. A traffic
light food labeling intervention increases consumer awareness of health
and healthy choices at the point-of-purchase. Prev Med. 2013 Oct;57(4):253-7. doi: 10.1016/j.ypmed.2013.07.001.
Thorndike
AN, Sonnenberg L, Riis J, Barraclough S, Levy DE. A 2-phase labeling
and choice architecture intervention to improve healthy food and
beverage choices. Am J Public Health. 2012 Mar;102(3):527-33. doi: 10.2105/AJPH.2011.300391. Epub 2012 Jan 19. Erratum in: Am J Public Health. 2012 Apr;102(4):584.
Levy
DE, Riis J, Sonnenberg LM, Barraclough SJ, Thorndike AN. Food choices
of minority and low-income employees: a cafeteria intervention. American Journal of Preventive Medicine. 2012;43:240-248.
なんだか今日のスピーカーはどちらも穏やか（
April 30 social norm
Topics:
- Social norms
- Strategic interactions, social preferences, and behavioral game theory.
Learning Objectives:
1. Define what is a “norm” – how is it different from customs
2. Explain the rational choice theory account of norms. Critique it.
3. Describe game theoretical accounts of norms (e.g. the norm for fairness and reciprocity).
Today’s Objectives:
1. Define a norm and its essential features.
2. Describe uses of subjective norms in health messaging.
3. Understand the norm of fairness.
Definition of norm
1. Specify what actions are regarded as proper/correct vs. improper/incorrect.
2.
Purposefully generated – people who maintain the norm see themselves as
benefiting from its being observed – or harmed by its being violated.
3. Enforced by sanctions (reward/punishment).
Wikiも参考になるかと
Why do we have norms?
1. To counteract externalities：
2. To coordinate actions：DCでは、エスカレータが立つのは右、
3. To enforce solidarity：
なぜnormを法律にしないか？
車の通るのが右というのは法律だがエスカレータは法律ではない
学生から出た意見
・危険すぎることは法律でそうでもないものはnorm
・法律にするには時間や手間などのコストがかかる
・法律にしても取り締まるのが大変
Economists’ view of norms
•Norms
= set of rules that coordinate expectations and thereby reduce
transaction costs in interactions that involve multiple equilibria.
•Either gold coins or sea shells could be used as money. There’s no “right” or “wrong” thing which could stand for currency.
•Or,
people can drive on the left (as in UK, New Zealand, Japan), or on the
right (USA, Europe). There is no “good” or “bad” side.
日本の東京と大阪ではエスカレーターで立つサイドが違う。
→よくわからないが発言するようにエンカレッジされた（
カワチ先生の答え
・東京はサムライで、大阪は商人の文化
・
など
Monetization
•Many norms are about money taboo.
•e.g. it’s acceptable in grad school to ask a friend to help with program code.
•How about paying them to do the analyses for your doctoral thesis?
•Monetization can destroy norms.
•Study conducted in 10 Israeli day-care centers.
•Norm = pick up your kids on time.
•At baseline: average ≍8 late pick-up of kids per week.
•N=6 centers selected for a new scheme in which parents were told they would be fined $3 if they were more than 10 minutes late.
•What happened next?
5th weekより罰金を開始
→軽い罰金を科すグループの方が遅刻が増えた！！
→つまり軽い罰金を科すことは「金払えば遅れてもいいんでしょ」
一方で罰金を１００ドルなどにすれば上記のような結果にはならな
ちなみに罰金中止（17th week）後も、介入群では遅刻は続いた。
•Financial incentives backfired.
•When money entered the equation, the social norm for “doing the right thing” became transformed into an economic transaction.
•Arriving late to pick up kid used to be perceived as being a “bad parent”.
•Imposing a fine destroyed the norm, and made it OK to arrive late so long as you paid the small fine.
2.Describe uses of subjective norms in health messaging.
Subjective norms
•[from psychology]: an individual's perception of whether other people think the behavior should be performed.
•Theory of Planned Behavior(TPB) – behavioral intentions are shaped in part by perceived peer norms.
What does TPB say?
•Behavioral intention is determined by attitudes, subjective norms, and perceived control over the behavior.
Normative Beliefs in TPB
•Subjective norms from peer group:
“My friends think I’m a wuss（臆病者、怖がり、意気地なし、根性なし、弱虫、
•Subjective norms from family:
“My parents disapprove of smoking”.
•Subjective norms from society:
“Everyone is against smoking”.
Many students over-estimate the prevalence of binge drinking on campus. Why?
->availability heuristic. 少数の激しく飲む学生が象徴的で印象に残るので（
•Over-estimating how much your friends drink can lead to your drinking too much. Why?
-> anchoring
•Problem
becomes mutually reinforcing because once you start binge-drinking,
your perception of your friends’ binge drinking becomes anchored on your
own behavior.
•Implication: interventions need to re-norm these misperceptions.
but
•Social norms marketing can sometimes backfire.
•Some studies suggest an increase in binge drinking following such campaigns.
Why?
Important distinction
•Descriptive Norms - perceptions of how other people are actually behaving, whether or not these are approved of.
•Injunctive Norms - behaviors which are perceived as being approved/disapproved by other people.
* it’s the distinction between norm= “the most commonly observed behavior” vs. norm = “behavior that is approved by others”.
Robert Cialdini’s (2003) experiment
oDisplayed alternate signs during 5 consecutive weekends.
oPlaced secretly marked pieces of petrified wood along visitor trails.
Effects of eye images (こういうのねhttp://www.staff.ncl.
3. Understand the norm of fairness.
Contrasting assumptions（授業で何度も出ている復習。
Homo economicus ← → Behavioral economics
Fully informed Boundedly rational (heuristics & biases)
Forward looking Bounded willpower (myopia)
Self-interested Boundedly self-interested
The Ultimatum Game
•Imagine a 2-person game in which you (“the proposer”) are given $10 to split with your partner.
•You are free to propose any kind of split – e.g. “$6 for me, and $4 for you”.
•You are under no obligation to be “fair”.
•But
once you make your offer, your partner is free to accept the deal or
veto it, in which case neither of you will receive any money.
日本語でいうところの「最後通牒ゲーム」のことですね
Class survey
Now, pretend you are “the proposer”. How much of the $10 would you offer to your partner (someone you’ve never met before)?
I offer $ __ out of $10
クラスでは相手に５ドルをあげるのが最多。
立場を逆にして学生に質問
Class survey
Now,
switch sides and pretend you are “the responder”. Circle the minimum
offer you would accept (your reserve price), below which you would veto.
I will accept:
$0 $1 $2 $3 $4 $5 $6 $7 $8 $9 $10
→３ドルくらいが多数。先行研究の結果、平均は２．
（古典的経済学で十分にトレーニングを受けていれば、０．
なぜこのようなこと（立場が変わると答えが変わる）になるか？
Answer:
Because not all humans are motivated by pure self interest (homo
economicus). Some people are prepared give up personal gains to maintain
the norm of fairness and to punish violators.
Dictator Game
•Players given $20 to split with an unknown stranger.
•They could either:
- be greedy and keep most of the money for themselves.
- be fair and split the money 50:50.
•Partner has no say (that’s why it’s called “dictator” game).
•76% of subjects went for the fair split.
•So far so good for the norm of fairness.
「独裁者ゲーム」ですね
↑面白い。これを読んでいて、税金が２０－３０%
Public Goods Game
•Trust game involving N > 2 people.
•Each subject given 10 tokens at the beginning of the session.
•A player can either:
(a) put token into private account, resulting in $1 of earnings for each token for that person; or
(b) put token into a public account, which is then doubled & split equally between the players at the end of the game.
•A selfish player would walk away with $10 at end of the game.
•But
if players cooperate (e.g. contributed all 10 tokens to public
account), they could double their money at the end of the game.
•Problem: Free-riders can take advantage of suckers in the game. Over repeated games, bad behavior drives out good behavior.
「公共財ゲーム」ですね。
Fehr & Gächter, AER 2000; 90:980-94
Without punishment, after 10 repeat sessions, cooperation drops to 10%.
With
the introduction of sanctions (allowing a player to apply “punishment
points” to other players which reduces their pay-off by 10%), compliance
approaches 100%.
以下には上記の論文の内容の解説も書いてあるっぽいです。
上記を読むと、
脳メカニズムの解説
The neural basis of the fairness norm compliance
•Trans-cranial direct-current stimulation (tDCS)
•9-volt battery, jumper cable to the brain
•< 1% of electrical discharge compared to ECT.
3 types of stimulation
•Anodal stimulation: positive (V+) stimulation that increases the neuronal excitability of the area being stimulated.
•Cathodal
(V-) stimulation: decreases the neuronal excitability of the area being
stimulated. (Used to treat psychological disorders that are caused by
the hyper-activity of an area of the brain).
•Sham
stimulation: used as control condition -- emits a brief current but
then remains off for the remainder of the stimulation time.
↓ これを応用した研究（読む価値ありそう、ある意味結構怖い研究）
Changing social norm complicance with nonivasive brain stimulation
Science, October 23, 2013
C.C.Ruff
PMID: 24091703
自分勝手な奴（特にわがままな金持ち？）
Implication
•tDCS does not make people “more moral”.
•Rather, it makes them more sensitive to the presence or absence of sanctions.
•Without credible sanctions, people are less likely to comply with norms.
•There
seems to be dedicated neural mechanism that enables us to flexibly
align our behavior with the possibility of social punishment.
Readings:
- James S. Coleman. The Demand for Effective Norms (chapter 10). In: James S. Coleman. The Foundations of Social Theory, Cambridge, MA: The Belknap Press of Harvard University, 1990, pp. 241-59.
- Jon Elster. Coleman on social norms. Revue Française de Sociologie 2003; 44(2):297-304.
- Karine Nyborg & Mari Rege. On social norms: the evolution of considerate smoking behavior. Journal of Economic Behavior & Organization. 2003; 52:323-340.doi: 10.1016/S0167-268(03)00031-3.
Incentives vs. Commitments
いまさらですが・・・（
Session 8 – April 16 (Wednesday)
Topic: Incentives vs. Commitments
Instructor: Professor Kawachi
Learning Objectives:
1. Distinguish between the use of incentives vs. commitments to influence behavior.
2. Describe the pros and cons of incentives vs. commitments.
3. Understand the principles of designing optimal incentives and commitments.
4. Describe the behavioral economics principles involved in “super-charging” incentives to alter behavior.
1. Distinguish between the use of incentives vs. commitments to influence behavior.
2. Describe the pros and cons of incentives vs. commitments.
3. Enumerate the principles of designing optimal incentives and commitments.
Incentives とCommitmentsの違いは？
・Incentives guide choice e.g. 25 cent tax on cigarettes
・Commitments disable choice e.g. $14 excise tax on cigarettes
JAPANESE DRUNK DRIVING LAWS
Driving Under the Influence BAC ≥ .08 Prison with hard labor up to 5 years or fine up to 1,000,000 yen.
Driving While Impaired, BAC .03 to .079 Prison with hard labor up to 3 years or a fine up to 500,000 yen
これだけではなく
・Serving alcohol to someone who subsequently gets a DUI or DWI :
Punished together with DWI driver.
・Riding as passengers in a vehicle operated by an intoxicated person:
Punished with DWI driver.
ここで(笑)がおこる。
実際に件数は減っているようだ
A weight loss pre-commitment device
•Reduced-strength version of orlistat (Xenical).
•Works by disabling lipase, thereby preventing the break-down of dietary fat while it's in your digestive tract.
•Approved for over-the-counter sale to adults 18 years and older.
Basically works like Antabuse
After fatty meal, users experience
Put Your Money Where Your Butt Is: A Commitment Contract for Smoking Cessation
RCT Design
•N=2000
smokers in the streets of Mindanao approached on spec & offered the
opportunity to sign up for a commitment contract to stop smoking.
•The contract required smoker to deposit money in a bank account (earning no interest) for 6 months.
•At 6 months, if he failed a urinary cotinine test, the smoker forfeited his money (donated to charity).
•2 Control groups:
a)Received wallet-sized “cue cards” showing frightening pictures of the health consequences of smoking,
OR
b) Nothing.
What Did They Find?
•11 percent of smokers who were offered the contract took it up.
•Average
client made a deposit every 2 weeks, and ended up committing 550 pesos
(USD $11) by the end of the 6 month contract period.
•≍ 20% of monthly income, or 6 months’ worth of cigarettes.
Percent passing urine test at 6 months (ITT):
Contract18.1%介入群
Cue cards15.3%脅す
Nothing12.4%
Pros and Cons
Incentives
•
An excise tax affects everyone regardless of their stage of change. (in
fact, raising the tax might nudge smokers to move to the next stage).
Commitment
• Low participation rate – contingent on smokers being in the preparation stage of behavioral change
(11% in Philippines study).
prochaska & diclemente 1983
Stage 1. Pre-contemplation
•Individual is not intending to take action in the foreseeable future (next six months).
•People may be in this stage because they are uninformed or tried & failed and become demoralized.
•People tend to avoid thinking about their behavior.
Stage 2. Contemplation
•People intending to change in the next six months.
•They are more aware of the pros of changing but are also acutely aware of the cons.
•This
balance between the costs and benefits can produce profound ambivalence
that can keep people stuck in this stage for long periods of time.
•i.e. behavioral procrastination.
Stage 3. Preparation
•Stage in which people are intending to take action in the immediate future, usually measured as the next month.
•They
have a plan of action, such as setting a quit date, recruiting social
support, talking to their physicians, or contacting a Quit Line.
Prev Med. 1995 Jul;24(4):401-11.
Distribution of smokers by stage in three representative samples.
Velicer WF1, Fava JL, Prochaska JO, Abrams DB, Emmons KM, Pierce JP.
Abstract
Objectives.
A key variable for the design of individual and public health
interventions for smoking cessation is Stage of Change, a variable which
employs past behavior and behavioral intention to characterize an
individual′s readiness to change. Reactively recruited samples distort
estimates of the stage distribution in the population because such
samples attract a disproportionate number of late-stage participants.
Three representative samples are described which provide accurate
estimates of the stage distribution in the population. These samples are
of adequate size to permit within-sample comparisons with respect to
sex, age, Hispanic or non-Hispanic origin, race, and education level.
The implications of using stage distribution as a tool for planning
intervention is discussed. Method. The first sample of 4,144 smokers was
from the state of Rhode Island and involved a random-digit-dial survey.
The second sample of 9,534 smokers was from the state of California and
involved a stratified random-digit-dial survey. The third sample of
4,785 smokers was from a total of 114 worksites located in four
different geographic locations. Results. The stage distributions were
approximately identical across the three samples, with approximately 40%
of the sample in Precontemplation, 40% in Contemplation, and 20% in
Preparation. The stage distribution was generally stable across age
groups with the exception of the 65 years and older group. Education
level did affect the stage distribution with the proportion of the
sample in Precontemplation decreasing as education level increased. In
all three samples, minor differences in stage distribution were related
to Hispanic origin and race, but the pattern was not consistent across
the samples. Conclusions. The pattern of stage distribution has
important implications for the design of interventions. Existing
interventions are most appropriate for the Preparation stage, but the
majority of the three samples were in the first two stages, resulting in
a likely mismatch between the smoker and the intervention. The
stability of distribution across age suggests that interventions that
are appropriately matched to stage can be applied across all age groups.
The differences found with respect to education, Hispanic origin, and
race can serve as a guide to the tailoring of intervention materials.
Implication
•About 80% of smokers are stuck in either pre-contemplation or contemplation stage.
•Only 20% are in preparation (i.e. planning to quit in next month).
•Most clinical interventions are targeted at preparation stage.
•As are pre-commitment devices.
•What’s the implication for incentives vs. commitments for reducing SES disparities in smoking?
プレコミットメントと課税はどちらが有効か？
→喫煙率が高いSESの低い人は低所得なので、課税がよい、と。
Incentives
• Revenue-generating, provided that drop in demand is < increase in taxes.
• Price elasticity of demand for cigarettes ≍ 8%.
Commitment
• Can be expensive (if carrots are big).
• At best, revenue-neutral, provided smokers commit their own money to quitting contracts.
Which do you prefer?
•Chance to win $10,000 reward for successfully quitting?
- or –
•Chance to lose $10,000 of your own money for failing to quit?
•Loss aversion may explain reason for low participation rates.
全員が前者に手を挙げた。しかしどちらが有効かと言えば、
HOW SHOULD WE DESIGN OPTIMAL INCENTIVES?
Case Study: Price subsidy on healthy foods in a supermarket
Am J Clin Nutr. 2010 Mar;91(3):736-47. doi: 10.3945/ajcn.2009.28742. Epub 2009 Dec 30.
Effects of price discounts and tailored nutrition education on supermarket purchases: a randomized controlled trial.
Ni Mhurchu C1, Blakely T, Jiang Y, Eyles HC, Rodgers A.
Randomized controlled trial of shoppers in 8 New Zealand supermarkets.
•N = 1,104 shoppers randomly assigned to one of 4 arms:
1) Price discount (12.5%) on healthier foods.
2) Tailored nutrition education
3) Discounts plus education
4) Nothing
Why target supermarkets?
65% of food expenditure in NZ
(compared with 33% in USA).
Why target food pricing?
•Cost cited as major barrier to consumption of healthier foods – especially for lower socioeconomic groups.
•Price discounts already used by retailers to boost sales.
Why tailored education（tailored educationとは？→例：マオリにはマオリ語、
•Systematic
review (15 trials, n = 20,809) showed tailored messages significantly
more effective in improving long-term diet than generic messages or no
education.*
•Tailoring can be to both individuals as well as cultural groups (e.g. ethnicity)
* Eyles & Ni Mhurchu, Nutr Rev, 2009;67(8):464-80.
SHOP trial design
•3-month baseline
→ 6-month intervention
→ further 6-month follow-up phase.
•Individualised electronic purchasing data used to tailor education & to measure effect of interventions.
•Main outcome measure: Change from baseline in % energy derived from saturated fat at 6 mo.
Price discount intervention
•Discount (subsidy) applied automatically at checkout to a list of 1,032 healthy food options:
–Generated from database of 12 months of retrospective electronic sales data from 6 supermarkets.
–The 1,032 selected food items were based on NZ National Heart Foundation criteria.
•Participants
sent list of discounted foods at start of 6 month intervention.
Reminder list sent at mid-point of intervention (3 months in).
•Why 12.5% discount?
- equivalent to not paying NZ Goods and Services Tax (GST).
Tailored nutrition education intervention
•Individually tailored using each shopper’s unique electronic food purchase data.
•Accompanied by recipes and suggfested portion sizes.
•Face-to-face group education also offered to ethnic minority (Maori and Pacific Islander) participants.
Effect of price discount
•Shoppers
receiving discount bought significantly more healthier food options at 6
mo. (1.7 lb. more per week than control group, on average).
•≍ 1.0 lb. more fruits/vegetables per week.
Intervention effects on food purchases
Difference in mean changes from baseline for price discount
and nutritional education interventions, kg/week
(Model adjusted for baseline nutrient values, ethnicity, income, age, sex)
→６か月後は、
However…
•At
6 mo. The difference in % energy from saturated fat（これがメインのアウトカムだったらしい）
for price discount group was -0.02% (95% CI: -0.40%, +0.36%) compared
to control group, p = 0.91.
•Corresponding change for tailored education group was -0.09% (95% CI: -0.47%, +0.3%), p = 0.66.
•? Substitution effect？
→他の商品でお金が残っていたので、
Summary
•No
effect of price discounts or nutrition education on saturated fat or
other nutrient outcomes. (But we already know it’s jolly difficult to
alter blood lipids through diet).
•Significant and sustained effect of price discounts on volumes of healthier foods purchased.
•No
effect of nutrition education on volumes of healthier foods purchased.
Combining education with price discount did not boost effect of subsidy.
Commentary from behavioral economics perspective
•SHOP trial only examined effect of a price subsidy. Why?
ジャンクフードに課税するのは現実には難しい。
•In conventional economics, subsidizing a health option is viewed as the same as taxing an unhealthy alternative.
•But behavioral economics says consumers view subsidies vs. taxes differently (asymmetry in price elasticity). Why? →loss aversion
Standard and Behavioral Economics Contrasted
Neoclassical economics
•Price increases (taxes) are just the flip side of price decreases (subsidies).
•The price elasticity of demand for a product is assumed to be symmetrical.
•If
the price of a can of Pepsi goes up by 50 cents, demand will drop by
5%, If it goes down by 50 cents, demand will rise by 5%.
Behavioral economics
•Loss aversion says that price elasticities will be asymmetric.
•Consumers will dislike price increases twice as much as the same price cut.
•Loss averse customers will reduce their consumption more when prices rise, compared to increasing purchases when prices fall.
同じくお金を出すにしても、ジャンクに課税するよりは、
To wit…
Tax increases work better than price subsidies.
and
since retailers will not raise prices voluntarily (due to competition),
there is a market failure for this form of incentive.
i.e. Government needs to do it.
HOW ABOUT USING COMMITMENT TO PROMOTE HEALTHY EATING?
Case Study: Voluntary pre-commitment device among grocery shoppers.
Psychol Sci. 2014 Feb;25(2):538-46. doi: 10.1177/0956797613510950. Epub 2014 Jan 3.
Healthier by precommitment.
Schwartz J1, Mochon D, Wyper L, Maroba J, Patel D, Ariely D.
Abstract
We
tested a voluntary self-control commitment device to help grocery
shoppers make healthier food purchases. Participants, who were already
enrolled in a large-scale incentive program that discounts the price of
eligible groceries by 25%, were offered the chance to put their discount
on the line. Agreeing households pledged that they would increase their
purchases of healthy food by 5 percentage points above their household
baseline for each of 6 months. If they reached that goal, their discount
was awarded as usual; otherwise, their discount was forfeited for that
month. Thirty-six percent of households that were offered the binding
commitment agreed; they subsequently showed an average
3.5-percentage-point increase in healthy grocery items purchased in each
of the 6 months; households that declined the commitment and
control-group households that were given a hypothetical option to
precommit did not show such an increase. These results suggest that
self-aware consumers will seize opportunities to create restrictive
choice environments for themselves, even at some risk of financial loss.
KEYWORDS:
health, rewards, self-control
PMID: 24390824 [PubMed - in process]
Ongoing field experiment in South Africa.
• One feature is a monthly 25% cash-back bonus awarded to
members’ credit card balances for healthy food purchases
during the past month.
• Even with this incentive, program remains under-utilized
How to boost success?
•n = 6,750 households already participating in trial were offered the chance to sign up for a pre-commitment device, viz.
•Put their monthly bonus on the line.
•If
their healthy food purchases did not increase by 5% over their
historical baseline for the next 6 months, they
forfeit（【他動】〔権利・名誉などを〕失う、喪失する、
緑が上昇したあたりが介入開始の時期
なんとなく介入開始に向けてグラフ全体が低下しているように見え
↑in % healthy foods occurred as a function of ↓unhealthy & neutral foods.
数値は小さいは有意な効果があったと。
How should we design optimal incentives?
The incentive must be “just the right size” -- neither too small, nor too big.
Using Incentives to Encourage Healthy Eating in Children
Field experiment in 15 elementary schools in Utah.（ユタ州らしい）
•Cluster randomization over 5 lunch periods, announced over school radio:
(1) receive 25¢ for eating one serving of F/V（フルール&ベジタブル）
(2) receive 5¢ for eating F/V
(3)
enter a raffle to win prize (value equal to total no. of kids who ate
their F/V that day multiplied by x25 – e.g. tennis racket)
(4) no incentive (control)
２５セントをあげるだけで、
親としては、いちいち説得するよりは、
Paying kids to eat their vegetables?!Sounds controversial…
Suppose you are the Superintendent of your School District…
WHAT MORE WOULD YOU LIKE TO KNOW ABOUT THE INTERVENTION?
・長期的な効果が知りたい→お金をやめた途端食べなくなりそう
・子どもがそのお金をなにに使ったか知りたい。
・実際本当にどれくらい食べたのか知りたい（
Q) did the financial incentive make “bad” social norm?
Old norm was 野菜を何も食べない事だったが、
Q)
４週間後には食べなくなっていました。
Q)子供たちは食べ物をサーブしただけで残して捨てていたか？
コントロール（４５％食べ残し）と比べて、２５セント、
Cost of providing rewards
•A quarter each day to each student ≍ $45 per year
•Labor costs of personnel ≍ $24 per child (based on two workers per school paid $8.50 per hour).
•TOTAL = $69 per child per year.
•Comparable to USDA Fresh Fruit & Vegetable Program ($50-75 per student、既に行われている政策らしい).
•But the incentive program cuts down waste!
What happens when incentives are too big?
Incentives & performance
•Performance incentives assume 2 things:
a) that incentives will increase motivation and effort;
•Sometimes,
this assumption backfires – e.g. monetizing the performance of a
behavior may paradoxically de-motivate (will say more about this in
lecture 10).
Incentives & performance
•Performance incentives assume 2 things:
a) that incentives will increase motivation and effort;
b) that increased motivation & effort will enhance performance.
•Paradoxically, when the reward is too high, motivation & effort are indeed greater, but performance can suffer.
•People choke.
What could make people choke under pressure?
1) Arousal. arousal【名】覚醒、目覚め、喚起、性的興奮
(Yerkes-Dodson Law, 1908).
What could make people choke under pressure?
1) Arousal.(Yerkes-Dodson Law, 1908).
2) Costs of shifting mental processes from “automatic” to “controlled”.
(e.g. focusing on your golf swing).
3)Narrowing of attention.
- can be detrimental for tasks that involve creativity & “thinking outside the box”.
4)Preoccupation with the reward itself.
- e.g. CEO bonuses on Wall Street.
•Hard to study incentives that are “too large”.
•NSF research budgets won’t allow it…
•except in India.
HarperCollins 2010←本
Review of Economic Studies (2009) 76, 451–469 0034-6527/09/00160451$02.00
2009 The Review of Economic Studies Limited
Large Stakes and Big Mistakes
N=87 residents in rural India.
•Participants invited to play 6 games involving memory, creativity & motor skills.
Structure of incentives
•Randomization to 3 incentive structures:
4 rupees
40 rupees ←ほぼ月収（１ドルで・・？？）
400 rupees* ←ほぼ年収（１０ドルで・・？？）
•Full payment if players reached “very good” level of performance, 50% payment if they reached “good” level.
*Rs
47.9 = US $1. Maximum incentive is close to all-India monthly average
per capita expenditure in rural areas. In the unlikely event that a
player reached top performance in all 6 games, she could earn roughly
half of annual income.
Preview of Lecture 11
•I will say much more about the topic of structuring incentives leveraging behavioral economics principles in lecture 11.
•…when we discuss behavioral economics & health policy.
•Optimal design of incentives involves choices about – size, timing, frequency, framing, duration.
８つスタンプをためたら１つ無料になるものと、
関連記事
How to structure commitment devices
1)Sticks tend to be more effective than carrots.
Why?
->loss aversion!
2)Commitments need to be monitored, enforced, and sanctions applied (if needed)
Principles of monitoring
1.You need to enlist a referee to check compliance.
i.e. announce to others that you are entering a commitment.
2.Not your spouse or significant other!
3.Not a fellow smoker!
4.Ideally, select more than one referee (In other words, leverage the power of peer support).
Commercial “Commitment Store”
コミットメント
•A “commitment store”.
•People register to enter into a commitment contract – they stake their own money & nominate referees to monitor progress.
•As of this week, the site lists 180,000 contracts involving $11 million on the line.
健康だけではなく、「論文一本書く」とかでもOK
日本でもやっている人がいるようだ http://angelux.jp/wpblog/?p=
Readings:
1. Volpp KG et al. Financial incentive-based approaches for weight loss. A randomized trial. JAMA 2008;300:2631-7.
2.
Volpp KG et al. A randomized controlled trial of financial incentives
for smoking cessation. NEJM 2009; 360:699-709.
3.
Xavier Giné, Dean Karlan & Jonathan Zinman. Put your money where
your butt is: A commitment contract for smoking cessation, Journal of
Economic Literature, December 2008.
Expectations – and why behavior interventions occasionally misfire
Session 10 – April 23 (Wednesday)
Topic: Expectations – and why behavior interventions occasionally misfire
Instructor: Professor Kawachi
Learning Objectives:
1. Understand how our expectations shape our subsequent experiences.
2.
Provide examples of how expectations can be manipulated to influence
choice (in either healthy or unhealthy directions).
3.
List and describe situations under which behavioral interventions can
misfire – substitution effects, behavioral compensation, “mere peanuts”,
moral licensing and “what the hell” effects.
人は先行する情報からの期待により食べ物に対する認識が変わる。
論文：The colr of odors_Gil Morrot_France_Brain and Language_2001
赤く色を付けた白ワインを、
論文：Physiology & Behavior_2007_Brian Wansink_Fine as North Dakota wine_Sensory expectations and the intake of companion foods
同じワインで、ラベルを変えるだけで、
論文：Food Quality and Preference_2005_Brian_how descriptive food names bias sensory perception in restaurants
同じものをサーブするにしても、
Red Beans and Rice
と表記するよりは
Traditional Cajun Red Beans and Rice
と記載したほうが、人は美味しいと思うらしい（
禁煙社会のための政策を進めた人：自身は喫煙者
Luther Terry (15 September 1911 – 29 March 1985)
BOOK: Eat, drink, and be healthy
この効果は？
一般的には、
・持っていること自体かっこ悪い
・味がまずいと感じる
Six reasons why behavioral interventions get sabotaged
1.Behavioral compensation
Definition: When people adjust their behavior in response to an intervention in order to maintain homeostasis.
Examples:
1.Changes in smoking topography (e.g. puff duration, puff volume) to maintain the same nicotine levels.
2.Installation of energy-efficient light bulbs leads to more energy use (Jevons’ paradox).
3.Increased
physical inactivity (following a bout of vigorous exercise) to maintain
the same overall energy expenditure level over a day.
頑張っても結局リバウンドするよね（長続きしないよね）
論
文：Body fat loss and compensatory mechanisms in response to different
doses of aerobic exercise - a randomized controlled trial in overweight
sedentary males
Mads
Rosenkilde , Pernille Landrock Auerbach , Michala Holm Reichkendler ,
Thorkil Ploug , Bente Merete Stallknecht , Anders Sjödin
American
Journal of Physiology - Regulatory, Integrative and Comparative
PhysiologyPublished 1 August 2012Vol. no. DOI:
10.1152/ajpregu.00141.2012
しかも頑張りすぎても結局他で代償されたりする（例：
2.Risk compensation
Definition:
when people adjust their behavior in response to the perceived level of
risk, becoming more careful where they sense greater risk and less
careful if they feel more protected.
•Although
usually small in comparison to the fundamental benefits of safety
interventions, it may result in a lower net benefit than expected.
Examples:
•Forestry workers taking more risks when wearing safety equipment (leg chaps).
•Eating more bacon after being put on a statin drug.（コレステロールを下げる薬を飲んでいるから、
•Engaging in risky sexual behavior after being put on antiretroviral therapy.（
→リスクをより適切に共有するにはどうしたらよいか？
この例として、NZの例（Hans Monderman (1945-2008), Dutch road traffic engineer）と東京の狭い道路の写真を例示あり。
ちなみに、日本はだれも自転車でヘルメットをかぶっていない・・
3.Substitution effects
[from economics] （いわゆる経済学でいうところの外部性ですね）An increase
in quantity demanded of a lower-priced good (a substitute) when price
is raised on the original good.
Example:
•Raising the price on soda (via tax) increases consumption of fruit juice.
Q. How is this different from behavioral compensation?
A. Behavioral compensation applies to the SAME behavior (e.g. maintaining energy balance or nicotine intake).
More
generally, by sending a message to the public about official
disapproval of a behavior (e.g. “too much fat in the diet”), consumers
may be driven to engage in a substitute behavior (consume unlimited
amounts of carbohydrates).
Example:
•Marantz
PR, Bird ED, Alderman MH. A call for higher standards of evidence for
dietary guidelines. Am J Prev Med. 2008 Mar;34(3):234-40.
他の例としては、国のガイドラインで、「
4.Moral licensing – ego depletion
[from
social psychology] Engaging in a virtuous behavior boosts our
self-image and “licenses” us to indulge in subsequent “bad” behavior.
何か「良いことをしている」と思うことが、
Example:
•Prius
drivers have among the highest rates of running through pedestrian
cross-walks ≍ 30% (Piff et al. “Higher social class predicts increased
unethical behavior”. PNAS, 2012; 109: 4086-4091).
Sidebar
Q. How is moral licensing different from moral hazard?
A.
Moral hazard applies to bad behavior whose costs are not entirely borne
by the agent, e.g. in some populations, the introduction of
antiretroviral therapy led to more sexual risk-taking (risk
compensation), resulting in increased STD.
BYOB_How
Bringing Your Own Shopping Bags Leads to Treating Yourself and the
Environment_Bryan K. Bollinger_Uma Karmarkar_Harvard Business School
リサイクルバッグ（いわゆるエコバッグ）を使っている人は、
ちなみに上記はHBSのレポートだそうで
「Reusable bags also increase probability of hedonic purchases by 1.41%.
•≍ additional $200 million in profits.」とか書いてあるところが、“素敵”ですね。
同じチョコでも、「low-fat」とラベルすると、
Wansink B & Chandon P. Can" low-fat" nutrition labels lead to obesity? Journal of Marketing Research 2006:605-17.
ということで、ラベリング（この商品は健康に良いです！）
5.“Mere peanuts” effect
Definition:
when the consequences of a behavior are delayed and intangible (“mere
peanuts”), consumers are apt to have self-control problems.
Examples
•Smoking one more cigarette is not going to cause cancer.
•Snack-size packaging of food are tempting to people on a diet.
論文：The effects of reduced food size and package size on the consumption behavior of restrained and unrestrained eaters
journal of consumer research 2008
パッケージがでかいとたくさん食べちゃうよ
6.“What the hell” effect
Definition:
When restrained eaters indulge in “forbidden” consumption and go over
their self-imposed daily limit, they reason that “the day is lost – what
the hell, let’s go all the way!”
Example:
Pizza experiment (Polivy et al. 2010)
Social influence
今日が終わると、Behavioral Economicsも残すところあと2回となりました。
Topic: Social influence
Instructor: Professor Kawachi
Learning Objectives:
1.
Describe key constructs and measures (reachability, degree &
closeness centrality, betweenness) in sociometric analysis.
2. Describe the “three degrees of influence rule”.
3. Critically analyze the threats to causal inference in empirical studies of social influence across social networks.
Three Network Ideas
1.The Small World Phenomenon (“the six-degree rule”)
2.Network Interventions
3.The Three Degrees of Influence Rule
The Small World Phenomenon
•Goal: To determine the network distance between any two people in the USA.
•Recruited
n=96 random people in Omaha Nebraska & instructed them to find a
way to deliver a letter to a stockbroker in Sharon, Mass. via a chain of
acquaintances.
Stanley Milgram (1933-1984)
“The small world problem”, Psychology Today 1967(2): 60-67.
Instructions in Milgram’s Experiment
•If you KNOW the target personally, send the letter directly to him.
•If
you DON’T KNOW the target personally, pass on the letter to one of your
personal acquaintances who is more likely than you to know this person.
•Add
your name to the roster at the bottom of this sheet so the next person
who receives this letter (and the PI) can keep track of who it passed
through.
The Six Degrees Rule
•The first letter reached the target in a few days via only two intermediaries.
•Note: this was not simply about looking up the target in the phone book. The last sender had to personally know him.
•The average degree of separation in experiment was 5.5.
•But only 18 letters eventually made it (19%).
Replication experiment
•In 2001, Columbia professor Duncan Watts repeated Milgram experiment on the internet.
•He recruited 24,000 senders to deliver an e-mail message to 19 targets in 157 countries.
•Average degree of separation ≍ 6.
Leskovec & Horvitz (2007)
•Examined dataset of 30 billion instant message conversations on Microsoft MSN Messenger conducted by 240 million users
More examples of small world networks
•Co-authorship networks: Any two researchers published in MEDLINE are separated by 4.4 co-authorship links (Newman, PNAS 2001).
•Neural
networks: Neurons in the brain of C. elegans worm are separated by 2.7
synapses (it has only 302 neurons and scientists have mapped ‘em all).
The Math behind Small World Phenomenon
•Suppose each of us has 100 friends.
•Then at one degree of separation, we are connected to 100 others.
•At 2 degrees, 10,000 people.
•At 3 degrees, 1 million.
•At 4 degrees, 100 million.
•At 5 degrees, 9 billion….
•But obviously, there’s redundancy – i.e. your friends know each other (there’s clustering).
Average path length = ln(N)/ln(K)
where N = number of nodes,
K = acquaintances per node.
•If N = 300 million (population of US), and each person knows 30 people, then APL = 5.7.
•If N = 6 billion (90% of world population, assuming 10% are too young to know anyone), and K = 30, then APL = 6.6.
→世界は狭い。
A Brief Primer of Social Network Analysis
Euler’s Königsberg Bridge Crossing Puzzle
ケーニヒスベルクの問題
Euler（←オイラーさんですね。http://ja.
•The starting point and ending point of a journey must have an ODD number of edges.
•A continuous path that crosses all bridges just once can only have one start point and one end point.
こういうのは英語で読んでもよくわからないので日本語で。
In general, a relation can be: Directed or Undirected Binary or Valued
1) undirected（
2) undirected, valued（強さの程度を数値で表現する）
3) directed, binary
4) directed, valued
Source: James Moody’s webpage (Duke Sociology): http://www.soc.
Centrality
The most intuitive notion of centrality focuses on degree (= number of ties an actor has).＝要するにいくつとつながっているかという指数
Source: James Moody’s webpage (Duke Sociology): http://www.soc.
Network Interventions
•A common type of network intervention is to identify opinion leaders to advocate for behavior change.
例)
•Who are opinion leaders?
- celebrities (hard to get them)
- self-identified (not reliable)
- expert-identified (e.g. ethnographers)
- sociometric (who receives most nominations)
→関係者に話を聞いてネットワーク図を描く。
例）
Valente et al. Substance Use & Misuse 2004 (39):1685-1712.
３，５，１５，１６がよい
何人に影響を与えているかが重要である
Degree centrality
1. Correlated with “popularity” in schools.→喫煙率の高い学校では、
2. Often the early adopters of innovations.
3. Recruited in peer education interventions.
But-
•Not always the most effective change agents.
•Leaders have vested interest in maintaining the status quo (e.g. smoking to be cool).
•In schools with high prevalence of smoking, they tend to be smokers. In low-prevalence settings, they tend to be non-smokers.
Betweenness centrality
Model
based on communication flow: A person who most frequently lies on the
shortest paths connecting others in the network (gate-keepers).
Source: James Moody’s webpage (Duke Sociology): http://www.soc.
Bridges (Liaisons)
•A person who links two or more otherwise disconnected groups.
•Bridges
are paradoxical in that they are weakly connected to groups, yet their
position gives them influence (potential brokers).
•Leaders are important within groups, but bridges are critical to diffusion between groups (both good & bad).
•Bridges may be less burdened than leaders; more amenable to change.
Isolates
•Individuals with no connections.
•Tend to be late adopters of innovations.
•Can be protective if they are quarantined from bad behavior.
Explanations for S-shape
1. Consumer heterogeneity.
Consumers are normally distributed with respect to their tendency to adopt an innovation.
2. Social influence via networks.
Diffusion
occurs via social networks. If diffusion occurred a-socially (i.e. each
individual has same probability of adopting), resulting curve would be
constant
“The Diffusion of Innovations”という本が、
Newly-acquired pre-cu;tural behavior of the atural troop of japanese monkeys on koshima islet←論文のタイトル
人間以外の動物で、どのように習慣・
What predicts speed of diffusion?
1. Characteristics of adopters – sociometric position.
2. Characteristics of the product – relative advantage over existing products, compatibility with consumer tastes, etc.
3. Characteristics of the network – network topology.
なぜ洗濯機の普及は遅かったか？
→冷蔵庫や電話やVCRは、それまでにないものであった。
Characteristics of innovations that speed diffusion
1) Relative advantage：Is the innovation superior to what existed before?
2) Compatibility: Does the innovation fit with the users’ values, norms, beliefs?
3) Complexity: Is the innovation easy to use?
4) Trialability: Can the innovation be tried before making a decision to adopt?
5)
Observability: Is the result of the innovation visible to others? Does
it give measurable results? 例）万歩計。誰にとっても見えやすい。フィードバックがかかる。
Taxonomy of Network Interventions
Intervention
1. Identify individuals based on network position
例）Find opinion leaders (agents with high in-degree centrality or betweenness centrality).
2. Structural alteration
例）
a) Add/delete nodes (quarantine)
b) Add/delete links (create bridges)
c) Rewire existing networks←同じ６つのnodeを経由する経路でも、
Thomas Valente (2012). Network Interventions. Science, 337: 49-53.
3.Segmentation
・Target groups (as opposed to individuals) for behavior change at the same time.
・SEGMENTATION：Detection of mutually exclusive groups through group-detection algorithm.
4. Induction
Activating peer-to-peer interaction to create behavioral cascades
•Tactic underlying “word of mouth” marketing (e.g. Amazon).
•Why social media might be more effective than mass communication (“Going viral”).
Thomas Valente (2012). Network Interventions. Science, 337: 49-53.
1.The Small World Phenomenon (“the six-degree rule”)
2.Network Interventions
3.The Three Degrees of Influence Rule
→Christakis氏の論文を紹介。興味ある人は、
Readings:
1.
Thomas W. Valente, Peggy Gallaher & Michele Mouttapa. Using social
networks to understand and prevent substance use: A trans-disciplinary
perspective. Substance Use & Misuse 2004; 39(10-12):1685-1712.
2. Nicholas Christakis & James Fowler, The spread of obesity in a large social network over 32 years. New England Journal of Medicine 2007;357:370-9.
3. Christakis N & Fowler J. The collective dynamics of smoking in a large social network.NEJM 2008;358:2249-58
4. Lyons R. The spread of evidence-poor medicine via flawed social-network analysis.Statistics, Politics, and Policy. 2011; 2(1): 1-26.
上記のChristakisのシリーズに対して、
The spread of evidence-poor medicine via flawed social-network analysis（たぶんフリーで読めます）
この論文は全部ちゃんと読んだ方が良さそうだなぁ（
Halo Effect for Bariatric Surgery
Collateral Weight Loss in Patients' Family Members FREE
Gavitt A. Woodard, MD; Betsy Encarnacion, MD; Joe Peraza, MD; Tina Hernandez-Boussard, PhD, MPH; John Morton, MD, MPH
Arch Surg. 2011;146(10):1185-1190. doi:10.1001/archsurg.2011.244.
肥満のために手術した場合、
Social Media and the Science of Health Behavior
Circulation. 2013;
Social Mediaは以下の4つの点で、
1.
Scale: The effects of social interactions qualitatively differ between
small vs. large groups. Small groups often unable to reach consensus
& generate norms.
2. Measurement: Ability
to observe behavior change in real time – e.g. sequence of adoption – is
important for determining tipping points and thresholds.
In social media, every interaction is time-stamped and recorded.
3. Structural control
4. Behavioral fidelity
@Science
The spread of behavior in an online social network experiment Damon Centola
“the behavior spread farther and faster across clustered lattice networks than across corresponding random networks.”
From Granovetter’s “strength of weak ties”
… to the “weakness of long ties”, when it comes to complex behavior change (including health behavior).
In conclusion
•A
social network approach shifts focus of intervention away from
individual characteristics toward people’s network position as well as
structural characteristics of their networks.
•Often
getting a group to change behavior can be easier than trying to change
people individually, esp. if the group can reinforce the message.
自分のノートには、いろいろFigがあるのですが、
デハ
最終回 Behavioral economics and health policy – what it adds and does not add
Session 15 – May 12 (Monday)
Topic: Behavioral economics and health policy – what it adds and does not add.
Learning Objectives:
1.
Critique the potential of applying behavioral economics to health care
policy (e.g. designing employee health incentives).
2. Debate the application of behavioral economics to public health policy (e.g. recent debates in UK government).
Readings:
- Volpp KG, Pauly MV, Loewenstein G, Bangsberg D. P4P4P: An agenda for research on pay-for-performance for patients. Health Affairs 2009; 28: 206-14.
- Marteau TM et al. Judging nudging: can nudging improve population health? BMJ2011;342:d228.
★ノートはここから★
Learning Objectives:
1. Describe how BE principles can be applied to “super-charge” incentives.
2. Discuss applications of BE for health care policy (P4P4P).
3. Summarize critiques of behavioral economics.
（今日は最終回なので最初は復習のようだ）
なぜカフェテリアのヘルシーラベリングは機能しないか？
・「健康的」というラベリング＝おいしい→子どもは避ける
・「健康的」というラベリング＝カロリーが少ないに違いない＝
・「健康的」というラベリング＝
カロリー表示の義務化はなぜうまくいかないか？
・そもそも人は読まない
・
・そもそも（ダメな）大人はそんなことには気にしない
（・少なく食べても後からまた食べちゃう＝backfire）
Descriptive norm campaign
Boomerang effect
ユーモアのあるキャラクターを使うと問題を真剣に考えないかも
Two BE principle used in “Deposit contract” as amoking cessation intervention?（
*Time inconsistent prefernces：将来の幸せより今の幸せ
*Loss aversion：一度獲得したものを失いたくない（
Drawback of “Deposit contract”は？
・Low participation誰もやりたがらない
学校で子供に野菜を食べさせるためにお金をあげることの良くない
・お金をあげるのをやめた途端に全く食べなくなる
・
What BE adds
●Assumptions of rational choice / expectancy-value theories
１）Agents are motivated by expected utility maximization.
２）Agents’ preferences are invariant, i.e. their judgments and choices do not flip-flop based upon irrelevant context effects.
３）Agents’ valuations of items are independent of possession.
４）Agents have consistent time preferences.
５）An agent’s utility is governed by purely selfish concerns – the theories do not take into consideration the utility of others.
●What behavioral economics adds
１）Agents decide based on visceral factors, such as the affect heuristic (lecture 2).
２）Agents’
preferences are susceptible to anchoring and decoy effects (lecture 3),
as well as the way choices are framed (lecture 4).
３）Owners
place a higher value on items than buyers (endowment effect), and the
pain of loss is more than twice the pleasure from gain (loss aversion)
(lecture 4).
４）•Agents display time-inconsistent, hyperbolic discounting (lecture 5).
５）
•Agents’ choices and actions are governed by social norms (lecture 8).
•An
agent’s actions seldom leave others unaffected. We are embedded in
social networks – we influence others, and in turn, others influence us
(lecture 9).
★たぶんここまでが復習
SIX CRITIQUES OF BEHAVIORAL ECONOMICS
1)Descriptive, rather than prescriptive.
•BE
provides a laundry list（〔いろいろなことを書き並べた〕長いリスト） of cognitive biases.
Theory doesn’t help us predict which bias will predominate in any given
situation.
•Theories need to make sharp predictions – in this regard, the rational actor model provides sharply testable hypotheses.
•By contrast, non-rational behavior permits deviations up or down.
•For example, the anchoring heuristic predicts that consumers will fail to adjust adequately from an initial guess.
•But BE won’t tell us where that guess comes from.
•BE proceeds by “post-hoc” explanations to fit the facts.
•Moreover, the number of post hoc explanations explodes when we let other heuristics enter the picture.
•By facilitating opposite predictions, BE often allows itself too many degrees of prediction.
David
Laibson & Richard Zeckhauser (1998). Amos Tversky and the ascent of
behavioral economics. J Risk and Uncertainty 16:7-47.
2)Limited generalizability of experimental findings.
•Too many experiments on college students (specifically economics & psychology majors)! この手の知見はやたらとアメリカの大学生を対象にしたものばかり
•More broadly, samples are almost entirely drawn from WEIRD societies – Western, Educated, Industrialized, Rich and Democratic.
↓
Behvaioral and brain sciences 2010 "The weirdest people in the world?" Joseph Henrich
“・96% of experimental samples come from countries with only 12% of the world’s population.
•In
JPSP, 67% of American samples (and 80% of samples from other WEIRD
countries) were composed solely of psychology undergraduates.”
“The muller Lyer illusion”（グーグルでイメージ検索かけて下さい、
Why? (speculation)
•Living in a "carpentered world," we have grown accustomed to seeing corners everywhere.
•The
brain is used to viewing corners and interpreting them as far and near,
and uses the same information to make size judgments.
•When
looking at the Müller-Lyer arrows, the brain interprets them as far and
near corners, and overrides the retinal information that says both
lines are the same length.
•In short, the Muller-Lyer illusion is a culturally evolved by-product of recurrent features in the local visual environment.
•But the illusion mainly occurs in WEIRD societies.
Mean offers in Dictator Games（DG）の結果も国も異なれば結果もかわる
Why?
•Higher offers in DG depends on the society’s degree of: a) market integration, and b) participation in a world religion.
•DG
may be capturing a specific set of social norms, culturally evolved for
dealing with money and strangers, that have emerged with the rise of
complex societies.
3)Works best for one-off behavior
e.g. defaulting people into a retirement savings plan when they go on the payroll.
Enrollment in 401(k) plans before & after automatic enrollment (AE) (Choi, Laibson, Madrian, 2001)
（上記では年金積立の参加を、
•Behavioral maintenance is much harder.
•…Especially behaviors that involve constant effort (e.g. daily exercise) or addictions (smoking).
•Nobody has (yet) cracked the long-term maintenance problem.
•Even one-off health behaviors (such as getting the seasonal ‘flu shot) takes additional effort.
PNAS, June 28, 2011
Using implementation intentions prompts to enhance influenza vaccination rates
Randomized trial of n = 3,272 employees to get their seasonal
‘flu shots.
無料のインフルエンザ33.1%がうけた
最初から日にちを決める→35.6%
最初から日にち＋時間も決める→37.1%
年金積立に比べれば効果は小さいけど、たいした手間暇なく４％
“Last mile” problem
Percent quit smoking at 6 months in World Bank study
Deposit Contract 18.1%
Education (cue cards) 15.3%
Nothing 12.4%
•Low-hanging fruit already plucked.
•Behavioral economics can boost success…
•But effect sizes are still modest…
•And it doesn’t suggest that we replace proven approaches.（
4)BE is unlikely to supplant existing theories and approaches
•BE is not a replacement for proven approaches such as excise taxes.
NY Time Op-Ed Economics Behaving Badly
Geoege Loewenstein and Peter Ubel(2010)
"It
seems that every week a new book or major newspaper article appears
showing that irrational decision-making helped cause the housing bubble
or the rise in health care costs.
Such insights
draw on behavioral economics, an increasingly popular field that
incorporates elements from psychology to explain why people make
seemingly irrational decisions, at least according to traditional
economic theory and its emphasis on rational choice. Behavioral
economics helps to explain why, for example, people under-save for
retirement, why they eat too much and exercise too little…But the field
has its limits. As policymakers use it to devise programs, it’s becoming
clear that behavioral economics is being asked to solve problems it
wasn’t meant to address. Indeed, it seems in some cases that behavioral
economics is being used as a political expedient, allowing policymakers
to avoid painful but more effective solutions rooted in traditional
economics…Behavioral economics should complement, not substitute for,
more substantive economic interventions. If traditional economics
suggests that we should have a larger price difference between
sugar-free and sugared drinks, behavioral economics could suggest
whether consumers would respond better to a subsidy on unsweetened
drinks or a tax on sugary drinks.But that’s the most it can do. For all
of its insights, behavioral economics alone is not a viable alternative
to the kinds of far-reaching policies we need to tackle our nation’s
challenges."
BEだけでPH上の問題が解決できるわけではなく、BEは、
5)BE is being used to promote a conservative political agenda
•In
the UK, the coalition government has promoted BE through its Behavioral
Insights Team based in the Cabinet Office, advised by Richard Thaler.
•See debate in British medical press:
- Bonell et al. “One nudge forward, two steps back”, BMJ 2011;342:d401.
- Bonell et al. “Nudge smudge: UK government misrepresents “nudge”. Lancet 2011; 377:2158-9.
- Marteau et al. “Judging nudging: can nudging improve population health?” BMJ 2011; 342: d228.
Is nudge an effective public health strategy to tackle obesity? NO_BMJ_2011 PMID: 21493668
“The
UK coalition government’s enthusiasm for nudge…dispenses with the
complexity of real life contexts and acknowledges only the immediate
proximal horizons of consumer choice.
“At a
stroke, policy is reduced to a combination of cognitive and “light”
environmental signals, such as location of foods within retail
geography.
“Nudge, along with the
responsibility deals [with the food & alcohol industry] is presented
as the alternative to regulation, or in the media jargon, the “nanny
state”.
“Our worry is that nudge becomes
collusion between the state and corporations to hoodwink consumers. At
least nannies are overt.”
6)Ethics of default options
例）臓器移植の例。
The upside of behavioral economics approaches
1.
If it doesn’t cost anything, boosting success by even a few percentage
points amounts to an important public health benefit. Just do it!
効果が小さくてもたいした手間じゃないならやらない理由は無い
2. Could help to
narrow health disparities (contrast to education-based approaches). High
SES people already help themselves. To the extent that low SES people
are more susceptible to decision errors, BE-based approaches will help
to close that gap. BEは健康格差を小さくできる（
休憩。
Applications of Behavior Economics to Health Policy
1. Health care policy
2. Public health policy
Behavioral economics application to health care policy
•Patient Protection and Affordable Care Act Section 2705
•Beginning
in 2014, employers may use up to 30% of total amount of employees’
health insurance premiums to provide outcome-based wellness incentives.
•Incentives
can take the form of “discount/rebate on premiums, or waiver of
cost-sharing mechanisms (deductibles, co-payments, co-insurance).”
Pay-for-Performance
P4P：いわゆる医療制度の政策誘導（
•Incentives targeted to providers (clinicians and hospitals) to improve quality of care.
P4P4P (Pay-for-Performance-for-
•Incentives targeted to patients to improve behavior.
Problem:
•Most insurance premiums bear no relationship to the individual policy-holder’s behavior.
•(No evidence that differential premiums affect health behavior).
How come differential insurance premiums are not very effective?
1. Delayed gratification – benefits of quitting smoking today won’t be reflected until next year’s premium adjustment.
2. Salience – nobody notices they are paying less for insurance.
3. Equity – lower-income individuals would end up with higher premiums. 貧困者ほど喫煙して肥満で運動しない結果保険料が高くなるという
Designing employee incentives using behavioral principles
Why not leverage employees’ decision errors to “super-charge” incentives?
Decision error
Present-biased preferences (myopia)
Leverage
Make rewards for healthy behavior immediate and frequent.
JAMA. 2008 Dec 10;300(22):2631-7. doi: 10.1001/jama.2008.804.
Financial incentive-based approaches for weight loss: a randomized trial.
Volpp KG1, John LK, Troxel AB, Norton L, Fassbender J, Loewenstein G. PMID: 19066383
N=57 obese individuals, randomized to:
- Lottery incentive program
- Control group
•Weight loss goal: 1lb. per week for 16 weeks
Immediate & frequent rewards
1. Participants in the lottery group eligible for a daily lottery prize if their weight was at or below their goal.
2.
Lottery provided frequent small payoffs (1 in 5 chance of winning $10)
as well as infrequent large payoffs (1 in 100 chance of winning
$100). ヤマダ電機のポイントがもらえるスロットを思い出すｗ
3. Results of lotteries texted daily to participants (immediate tangible feedback).
4.…although actual money was paid out only if they met their goal at monthly weigh-in (regret aversion).
Why not leverage employees’ decision errors to “super-charge” incentives?
Decision error
Regret aversion “「後悔回避：regret aversion」とは、現在および将来における「後悔を嫌い、
Leverage
Tell people the total $$ amount they would have won if they had maintained weight loss at monthly weigh-in.
Why not leverage employees’ decision errors to “super-charge” incentives?
Decision error
Mental accounting “[メンタル・アカウンティング(mental accounting、心の会計)とは、
シカゴ大学の行動経済学者Richard Thaler氏は、次のような問いを設定した。
映画を見に行こうとして、
この問いに対して「イエス」(もう一度買う)と答えた人は46％
料金が1人10ドルの映画館に、映画を観に行きました。
こちらの質問では、88％
Thaler氏は次のように説明する。
脳がメンタル・アカウンティングを行なうために、
Thaler氏の別の実験では、「
Leverage
Separate reward is more motivating than a discount on insurance premium.
Mental accounting (Thaler, 1980)
•Individuals divide their current and future assets into separate, non-transferable buckets.
•e.g.
starting a savings account for the next family vacation at the same
time as carrying a large balance on a high APR credit card.
•By
the same token, people put “winning $250 on a smoking cessation
lottery” in a separate mental account from “receiving a $250 deduction
on a health insurance premium”.
•Note the connection to Framing (lecture 4).
How to super-charge incentives
Incentive Features -Lesson Learned
１）Timing -Should be frequent, and preferably soon after the behavior occurs (i.e. not wait till next year’s premium adjustment).
２）Size -If
the incentive has small value, turn it into a lottery to take advantage
of variable reinforcement schedule as motivating factor.
３）Framing -Same amount of reward (e.g. $300) feels different if it’s presented as a bonus vs. a discount on premium.
４） Carrot
or Stick? -People prefer carrot > stick.(Loss aversion predicts that
sticks may be more motivating, but Volpp study found no difference)
Issues in scaling up
1）Long-term effects of incentives?
Do incentives reduce intrinsic motivation?
Once rewards are removed, do behaviors rebound?
What’s optimal duration of rewards to make behavior change stick?→
2）Should we reward Δ behavior or Δ outcomes?
Outcomes are easier to verify (e.g. Δ weight vs. Δ diet).
But subjects may feel cheated if Δ behavior does not lead to Δ outcome.
Sometimes
it makes more sense to incentivize Δ behavior, e.g. attending smoking
cessation program (as opposed to quitting cold turkey).
3）Who pays?
In
high turn-over private insurance market, it’s most cost-effective to
focus on incentives with short-term payoffs (e.g. asthma management,
smoking cessation).
For long-term
insurance providers (Medicare, VA), it may be cost-effective to focus on
long-term payoffs (e.g. blood pressure control, weight loss).
For programs with a high public health rationale (adherence to medication in patients with MDR-TB), government could step in.
Volpp KG et al. (2009). Health Affairs 28(1):206-14.
Resistance to P4P4P
1）Weak evidence base
2011
Cochrane review of 19 RCTs of smoking cessation concluded that with the
exception of one study (by Volpp et al.), there was no clear evidence
of long-term (> 12 months) success rate with financial incentives.
•However, most studies were small and did not have sufficient power to detect increase in long-term quit rates.
•The size of incentives was generally too small – as low as $10 in some studies.
A randomized, controlled trial of financial incentives for smoking cessation. Volpp KGら,PMID: 19213683、N Engl J Med. 2009
たぶん＠GEの話らしい。個人的にはこういうのやりたい。
N=878 employees in firm randomized to information about smoking cessation program, or information + financial incentives.
•$100 for completing smoking cessation program.
•$250 for biochemically confirmed cessation at 6 months.
•$400 for confirmed cessation at 12 months.
•i.e. maximum of $750 reward.
"Conclusions：In
this study of employees of one large company, financial incentives for
smoking cessation significantly increased the rates of smoking
cessation."
How much do you need to pay smokers to quit?
A. Kim et al. (2011). JOEM 53(1): 62-67.
行動変容のステージの違いで必要というお金の額が違う
Resistance to P4P4P
２）High turnover rates
Turnover
in private insurance markets (and among employees) is high, meaning
that savings in future medical spending is likely to accrue to someone
else.
３）Fairness concerns （自力で禁煙した人は、
Many
people favor penalty programs because they find it distasteful to
reward people for doing what they should be doing anyway in their own
self-interest.
In the Volpp study, once the company decided to scale up the program, feedback from nonsmoking employees resulted in replacement of the $750 reward with a $625 penalty for smokers.
Nonsmokers
believed that their colleagues shouldn’t be rewarded for “something I
did myself without any reward”. The anticipated difficulty of collecting
money from smokers who didn’t quit resulted in the penalty being
applied to higher health insurance premiums. Reward systems can be made
more salient by being segregated from paychecks, but penalties
cannot. But rewards outside of paycheck would also be taxable.
BEHAVIORAL ECONOMICS IN PUBLIC HEALTH POLICY
”Thomas Frieden, “A Framework for Public Health Action: The Health Impact Pyramid”, AJPH 2010; 100: 590-5.”の三角形の図
Two applications of BE in public health policy
1.Tobacco “end game”
2.HPV vaccination
1.Tobacco “end game”
Plos Medicine Debate "the case for a smoker's license"
The Argument
• Advanced countries ought to be planning for a tobacco “end game”.
• Future tobacco sales should be done via a “tobacco user’s license”, using Smart Card technology（日本のタスポって有料？
A Tobacco Smart Card?
•All smokers must register (and pay license fee).
•Retailers can only sell to Smart Card users (achieving more reliable enforcement of illegal sales to minors).
•Smokers pre-commit to maximum daily consumption (e.g. 20 per day). Incentive: lower license fee for lower daily consumption.
•Can enroll registered smokers for cessation interventions – e.g. Quit to Win lotteries.
２．HPV vaccine
•Recommended for all girls & boys ages 11-12.
•Could prevent 70% of all cervical cancer & 90% genital warts.
•Yet uptake rates remain very low (≍ 1/3 of girls aged 13-17 have been fully vaccinated).
•What’s the barrier?
Integrating epidemiology, psychology, and economics to achieve HPV vaccination targets.
Basu S, Chapman GB, Galvani AP.
Proc Natl Acad Sci U S A. 2008 Dec 2;105(48):19018-23. doi: 10.1073/pnas.0808114105. Epub 2008 Nov 17.
PMID: 19015536
Behavioral economics of boosting HPV vaccination uptake
•Problem is not lack of perceived susceptibility.
•Problem is not lack of perceived vaccine efficacy.
•Problem
is perceived risk compensation (“The vaccine will drive my child into
sexual frenzy as a result of feeling “protected”).
= affect heuristic (since there is no evidence that such behavioral consequences will result).
0 件のコメント:
コメントを投稿