[ ] No comparison ideal, then show R1 in no info is the same in R1 and leave out R1, do another treatment with Irrelevant Info and show it's the same, perceived obj about consistency more likely to influence results than other demand effects, delib treatment varies across treatments (can't compare -- would have to be the same), demand should be about positive effect on wtp (pt is objective doesnt matter), concerned that strips smaller effect size - then choose just info or just discount and still call it policy interactions
[ ] - Dont expect cost wtp to be diff between cost and both treatment. If this is the case can argue that delib should minimize interaction effects
[ ] - Tell them to choose the answer that makes the most sense to them
[ ] - Prioritize demand effect treatment
[ ] - Test no info without baseline as well for pretest and pilot?
[ ] - Price reduction also imp
[ ] - I expect anchoring
[ ] - People don't want to reviss
[ ] - Delete deliberation only
[ ] - Anchored by design
[ ] - Bad test for anchoring. Good for something else
[ ] - Are there interactions when delib only r3 is used instead of both r3? Can't use r2 or r1
[x] - Confusion about how led wtp translates to choices
[x] - For all led, just tell them upper bound is budget. We don't know upper bound! But can't spend more money than they have. Because lower = budget - 1.
[x] - Ask about confusion in verb prot
[ ] - Give them the answer vs get them to reflect on it
[ ] - Analogous to getting them to reflect on mistakes
[ ] - Deliberation without reflection on mistakes could be a between treatment
[ ] - 2 demand effect treatments - placebo test for deliberation and the other for high LED WTP
[ ] - Doing us a favor - not true. Deception
[ ] - Demand effect depends on perceived obj
[ ] - Demand - doesn't tell you there isn't demand effects, just tells you that it was constant and didn't affect treatment effect estimates
[ ] - Make sure r1 intro demand
[ ] - Point is if obj is constant, te unaffected
[ ] - Iwtp is leq sum and higher than normal
[x] - Replace I chose with I believe
[ ] Delib or choice reconsid?
[x] You can use total to guide your answers by comparing it to your LED WTP. All revisions on same page
Activate to edit the task name
[x] Endorse - make sure I say there are no right answers
[x] Ask them to confirm R3 selections - ask everyone (discount, no revise, prep on R2, ask them to confirm - ask them to look at them again; discount, recise, ask them to confirm - chance for discount framing to enter again)
[ ] Keep certainty for pre-test, delete for pilot?
[ ] **In pre-test, keep certainty for half . Certainty should reduce demand effect (if they realize they are uncertain, useful, if they want to tell me they are more certain than they really are then not useful)
[ ] Check which product was chosen for displaying budget
[x] Record number of people who went back after seeing certainty
[x] Do no certainty for just one condition and treat as separate cond
[ ] For example, if the price of A was and B was, and you chose B then.. make sure B is higher then A. Randomize whether smart is higher or trad is higher. Then get rid of examples elsewhere
[x] Add in calcs - ques about calculating bonus payment
[x] Cert - if selected all A, if selected all B, if not either of those
[ ] Only put the ex in cert, skip ex in payment details
[x] Tell them they can go back in cert and don't make it a question
[ ] Put in obj - going to look weird. More info (both cost and emissions info) decreases demand relative to less info. Discounts decrease demand when information is provided. Info decrease demand when discounts are provided and the opp
[x] No right or preferred with endorse
[x] Anytime they compare lhs or rhs of ineq, remind them about other attr
[x] **In obj - limit how many they can pick
[x] Mention signs in obj**
[ ] Make sure second round with robustness can rule out large effect sizes - select treatments
[ ] Choice reconsid in educ lit
[ ] Delib treatments - separate treatments
[ ] Do another delib only treatment with discount ?
[ ] Test another treatment - info, choices, delib, choices (like delib only)
[ ] Irrelevant info adds noise that isn't there in other treatments... noisy vs less noisy worse than less noisy and less noisy
[ ] Do demand effect treatment if there are doff in TE across diff perceived objectives
[ ] Delib - use your cost and emissions wtp to revise answers to increase led wtp. And thrn some for decrease. Bound the effect
[ ] **Demand - do us a favor if you revise and do us a favor if you don't revise
[ ] Demand - no right ans but you would do us a favor vs just you would do us a favor. Right but preferred**
[ ] 2 options - pay X or pay Y, receive 25 in five, reduce emissions by in five: think of it as a bundle
[x] Put I don't know as objective
[ ] Test whether I have guided everyone to be close to cost wtp! Make sure I remind them@of other attributes in attributes revise led revise, when I show inequality
[ ] Make sure I didn't decrease salience by checking attribute importance q - if present in r3"2, should be present in r3. Make sure all three rounds are there. List all other attributes as well? Remind them of other attributes each time I mention cost, etc.
[ ] Tell them there should be no relationship. Tell them they should not revise. Then tell them there should be a relationship and they should revise -- these are the bounds
[x] Put no relationship first
[ ] Secondary treatments: Demand, Delib Only, Both x Price Reduc, Cost and Irr,
[ ] Look through all downloadable instr
[ ] Demand
only revise LED WTP - don't say anything about attributes, get them to revise about the relationship being the sum. before that . Several demand effects - demand effect won't cleanly lead me to the hypothesis that I want. Address the one people are concerned about and see if that changes results. Might change the results if both info is smaller and people increase it -- counteracting interaction effects. Say smaller than it normally would. Separate objective to info/discount on demand vs. deliberation on demand (people might think it is about the effect of deliberation on demand). At least now the perceived hypothesis is not wrong. Weird to tell them about different treatments
[ ] electricity prices in cost savings more detailed info
[ ] ask for mailing addresses so we can send you the bonus payment and the lottery without asking for anything else
[ ] Learning effects anchoring vs objective !! Might be okay if i take the difference. Introduced noise
[x] IRB form
A statement that identifiers might be removed from the identifiable private information or identifiable biospecimens and that, after such removal, the information or biospecimens could be used for future research studies or distributed to another investigator for future research studies without additional informed consent from the subject or the legally authorized representative, if this might be a possibility
Plans for retention of identifiers associated with the data
Methods for safeguarding data (such as, encryption or limited access to identifiable data).
The protection of confidentiality occurs after the data is collected and in the researcher’s possession.
Subjects should be informed during the consent process of the methods that will be used to maintain the confidentiality of their identifiable information and the possible risks of disclosure of this information outside of the research study.
Keep sensitive and identifiable data in encrypted files on a password protected hard drive.
aunarang
commented
2 years ago
Created 8/18/22
[x] Check Round intros for no info
[ ] No comparison ideal, then show R1 in no info is the same in R1 and leave out R1, do another treatment with Irrelevant Info and show it's the same, perceived obj about consistency more likely to influence results than other demand effects, delib treatment varies across treatments (can't compare -- would have to be the same), demand should be about positive effect on wtp (pt is objective doesnt matter), concerned that strips smaller effect size - then choose just info or just discount and still call it policy interactions
[ ] - Dont expect cost wtp to be diff between cost and both treatment. If this is the case can argue that delib should minimize interaction effects
[ ] - Tell them to choose the answer that makes the most sense to them
[ ] - Prioritize demand effect treatment
[ ] - Test no info without baseline as well for pretest and pilot?
[ ] - Price reduction also imp
[ ] - I expect anchoring
[ ] - People don't want to reviss
[ ] - Delete deliberation only
[ ] - Anchored by design
[ ] - Bad test for anchoring. Good for something else
[ ] - Are there interactions when delib only r3 is used instead of both r3? Can't use r2 or r1
[x] - Confusion about how led wtp translates to choices
[x] - For all led, just tell them upper bound is budget. We don't know upper bound! But can't spend more money than they have. Because lower = budget - 1.
[x] - Ask about confusion in verb prot
[ ] - Give them the answer vs get them to reflect on it
[ ] - Analogous to getting them to reflect on mistakes
[ ] - Deliberation without reflection on mistakes could be a between treatment
[ ] - 2 demand effect treatments - placebo test for deliberation and the other for high LED WTP
[ ] - Doing us a favor - not true. Deception
[ ] - Demand effect depends on perceived obj
[ ] - Demand - doesn't tell you there isn't demand effects, just tells you that it was constant and didn't affect treatment effect estimates
[ ] - Make sure r1 intro demand
[ ] - Point is if obj is constant, te unaffected
[ ] - Iwtp is leq sum and higher than normal
[x] - Replace I chose with I believe
[ ] Delib or choice reconsid?
[x] You can use total to guide your answers by comparing it to your LED WTP. All revisions on same page Activate to edit the task name
[x] Endorse - make sure I say there are no right answers
[x] Ask them to confirm R3 selections - ask everyone (discount, no revise, prep on R2, ask them to confirm - ask them to look at them again; discount, recise, ask them to confirm - chance for discount framing to enter again)
[ ] Keep certainty for pre-test, delete for pilot?
[ ] **In pre-test, keep certainty for half . Certainty should reduce demand effect (if they realize they are uncertain, useful, if they want to tell me they are more certain than they really are then not useful)
[ ] Check which product was chosen for displaying budget
[x] Record number of people who went back after seeing certainty
[x] Do no certainty for just one condition and treat as separate cond
[ ] For example, if the price of A was and B was, and you chose B then.. make sure B is higher then A. Randomize whether smart is higher or trad is higher. Then get rid of examples elsewhere
[x] Add in calcs - ques about calculating bonus payment
[x] Cert - if selected all A, if selected all B, if not either of those
[ ] Only put the ex in cert, skip ex in payment details
[x] Tell them they can go back in cert and don't make it a question
[ ] Put in obj - going to look weird. More info (both cost and emissions info) decreases demand relative to less info. Discounts decrease demand when information is provided. Info decrease demand when discounts are provided and the opp
[x] No right or preferred with endorse
[x] Anytime they compare lhs or rhs of ineq, remind them about other attr
[x] **In obj - limit how many they can pick
[x] Mention signs in obj**
[ ] Make sure second round with robustness can rule out large effect sizes - select treatments
[ ] Choice reconsid in educ lit
[ ] Delib treatments - separate treatments
[ ] Do another delib only treatment with discount ?
[ ] Test another treatment - info, choices, delib, choices (like delib only)
[ ] Irrelevant info adds noise that isn't there in other treatments... noisy vs less noisy worse than less noisy and less noisy
[ ] Do demand effect treatment if there are doff in TE across diff perceived objectives
[ ] Delib - use your cost and emissions wtp to revise answers to increase led wtp. And thrn some for decrease. Bound the effect
[ ] **Demand - do us a favor if you revise and do us a favor if you don't revise
[ ] Demand - no right ans but you would do us a favor vs just you would do us a favor. Right but preferred**
[ ] 2 options - pay X or pay Y, receive 25 in five, reduce emissions by in five: think of it as a bundle
[x] Put I don't know as objective
[ ] Test whether I have guided everyone to be close to cost wtp! Make sure I remind them@of other attributes in attributes revise led revise, when I show inequality
[ ] Make sure I didn't decrease salience by checking attribute importance q - if present in r3"2, should be present in r3. Make sure all three rounds are there. List all other attributes as well? Remind them of other attributes each time I mention cost, etc.
[ ] Tell them there should be no relationship. Tell them they should not revise. Then tell them there should be a relationship and they should revise -- these are the bounds
[x] Put no relationship first
[ ] Secondary treatments: Demand, Delib Only, Both x Price Reduc, Cost and Irr,
[ ] Look through all downloadable instr
[ ] Demand
only revise LED WTP - don't say anything about attributes, get them to revise about the relationship being the sum. before that . Several demand effects - demand effect won't cleanly lead me to the hypothesis that I want. Address the one people are concerned about and see if that changes results. Might change the results if both info is smaller and people increase it -- counteracting interaction effects. Say smaller than it normally would. Separate objective to info/discount on demand vs. deliberation on demand (people might think it is about the effect of deliberation on demand). At least now the perceived hypothesis is not wrong. Weird to tell them about different treatments
[ ] electricity prices in cost savings more detailed info
[ ] ask for mailing addresses so we can send you the bonus payment and the lottery without asking for anything else
[ ] Learning effects anchoring vs objective !! Might be okay if i take the difference. Introduced noise
[x] IRB form
A statement that identifiers might be removed from the identifiable private information or identifiable biospecimens and that, after such removal, the information or biospecimens could be used for future research studies or distributed to another investigator for future research studies without additional informed consent from the subject or the legally authorized representative, if this might be a possibility
Plans for retention of identifiers associated with the data Methods for safeguarding data (such as, encryption or limited access to identifiable data). The protection of confidentiality occurs after the data is collected and in the researcher’s possession.
Subjects should be informed during the consent process of the methods that will be used to maintain the confidentiality of their identifiable information and the possible risks of disclosure of this information outside of the research study.
Keep sensitive and identifiable data in encrypted files on a password protected hard drive.
[x] Fix placeholders
[ ] capture how many people went back
[ ] add in dollar signs for WTP cost
[x] work on branch logic for displaying WTP