LEAD Lab-In-The-Field PDF - IPA Endline Training, January 2025

LEAD Lab-In-The-Field IPA Endline Training, January 2025 John Egyir Roadmap 1. Introduction and research question 2. Experimental method and sample 3. Instruments overview 4. Administering the instruments Introduction  What drives parenting style choices? Parental style: parenting behaviors like...

LEAD Lab-In-The-Field IPA Endline Training, January 2025 John Egyir Roadmap 1. Introduction and research question 2. Experimental method and sample 3. Instruments overview 4. Administering the instruments Introduction  What drives parenting style choices? Parental style: parenting behaviors like discipline practices and showing affection to their children (harsh discipline vs warmth) Why is this question important? Harsh discipline and low affection has been shown to decrease children’s skills and well-being Parenting and family programs aim to shift these practices  Crucial to know why parents use them to design these programs more effectively For example: Parents could choose harsh discipline because they have little time or are stressed Or they might belief that it is an effective strategy  Different elements to target by parenting programs Experiment and sample Experiment & sample Method: Lab-in-the-field experiment Elicit beliefs and costs around some parenting behaviors with hypothetical situations/scenarios: Hold certain information constant and vary characteristics Strategy: Hypothetical scenarios about children, their parents, and their environment We test: How much do caregivers believe parenting styles affect future outcomes for their children? How parenting styles influence the skills and time spent with the child? Do caregivers make different choices under stress/with less time available? Do caregivers find certain parenting styles particularly costly or difficult to use? Sample: LEAD caregiver sample Experiment & sample Your work is key for the success of the LIF project! Why? - Since LIF is embedded in QP4G-LEAD: We can use the available panel data you collect To learn how parents make parenting style choices and how this vary based on previously collected characteristics (e.g., stress, parenting behavior, etc.) It helps us to find mechanisms and evaluate whether the parenting training program (ie., LEAD) worked or not Super important to keep the same QP4G-LEAD sample that we interviewed over time Tracking is key! – losing too many caregivers will affect LIF sample and chances of using the panel data Also, as caregivers are asked to stay a bit longer for the LIF, it is important to make them stay for this additional module Instruments Overview LIF instruments (I): intro and consents Introduction after main questionnaire Invite caregiver to stay a bit longer. Participation is voluntary but they are incentivized to stay for an additional half an hour to respond to the LIF module with additional 20 cedis airtime Consent Ask the consent with the consent form of the LIF LIF instruments (II): belief scenario Belief scenario (i) Ask caregivers how much income they think a hypothetical child will earn later, and (ii) If the child will care for the parents when they are elderly Two (2) Situations (about the described child to ask) with 2 different time investment amounts (8 in total, for 2Qs) OR 2 different skill levels (8 in total, for 2Qs) Additional question to test the certainty of responses Administering Instruments: Belief scenario Belief scenario Description of a child, his/her characteristics (i.e., initial skill level and amount of time received from parents) and how the parents treat him/her We want to know, based on the descriptions and different situations, what are parents’ ideas about how much this child will earn and care for their parents in the future. Belief scenario: how it works in practice when initial skills are given (Route A) You (as an Enumerator) can open a questionnaire with a description of a child, how the parents treat him/her, and his/her characteristics (e.g., High initial skill or Low initial skill) Then ask about: Caregiver’s predicted income for child when child is 30 years-old Caregiver’s opinion on likelihood of the child taking care of parents or supporting younger siblings later Repeated in four ways: Variation: time spent with the child (amount) + parenting style described Route A: Fixed initial skill level See tablet: Intro text first – show on tablet/screen Child’s initial skill level Situation 1: [CHILD A]’s parents set high expectations for [him/her] and do not praise [him/her] often when [he/she] does well. They do not allow [CHILD A] often to express [his/her] opinion. If [CHILD A] misbehaves [his/her] parents punish [him/her] without an explanation. If they punish [him/her], they do that by shouting, scolding, or using other rigid forms of discipline. Situation 2: Now imagine the same [CHILD A] that you see here. The only difference with the previous situation is the way parents interact with [him/her]. In this new situation, [CHILD A]’s parents set high expectations for [him/her] and praise [him/her] often when [he/she] does well. They allow [CHILD A] often to express [his/her] opinion. If [CHILD A] misbehaves [his/her] parents punish [him/her], [her/his] parents explain the reason for the punishment and how to do better next time. If they punish them, they do that by withdrawing privileges, sending [him/her] to the room or giving extra household chores  Same child in both situations  Variation: Situation Order is randomized. Text in RED changes Route A: After each situation Ask caregiver about: Income for described child: As monetary value (in today’s Ghana Cedis) Care for parents when they grow old for described child: Likelihood to take care of parents and siblings Explained as concept with counters (scale 0-10) – more on slides 19 Repeated for: High amount of time spent with the child Low amount of time spent with the child Belief scenario: how it works in practice when time amounts are given (Route B) You (as an Enumerator) can open a questionnaire with a description of a child, how the parents treat him/her, and his/her characteristics (e.g., High amount of time received or Low amount of time received) Then ask about: Caregiver’s predicted income for child when child is 30 years-old Caregiver’s opinion on likelihood of the child taking care of parents or supporting younger siblings later Repeated in four ways Variation: initial skill level of the child + parenting style described Route B: Fixed amount of time received See tablet: Intro text first Amount of time child receives from parents Situation 1: [CHILD A]’s parents set high expectations for [him/her] and do not praise [him/her] often when [he/she] does well. They do not allow [CHILD A] often to express [his/her] opinion. If [CHILD A] misbehaves [his/her] parents punish [him/her] without an explanation. If they punish [him/her], they do that by shouting, scolding, or using other rigid forms of discipline. Situation 2: Now imagine the same [CHILD A] that you see here. The only difference with the previous situation is the way parents interact with [him/her]. In this new situation, [CHILD A]’s parents set high expectations for [him/her] and praise [him/her] often when [he/she] does well. They allow [CHILD A] often to express [his/her] opinion. If [CHILD A] misbehaves [his/her] parents punish [him/her], [her/his] parents explain the reason for the punishment and how to do better next time. If they punish them, they do that by withdrawing privileges, sending [him/her] to the room or giving extra household chores  Same child in both situations  Variation: Situation Order is randomized. Text in RED changes Route B: After each situation Ask caregiver about: Income for described child: As monetary value (in today’s Ghana Cedis) Care for parents when they grow old for described child: Likelihood to take care of parents and siblings Explained as concept with counters (scale 0-10) – more on slides 19 Repeated for: High initial skill level of child Low initial skill level of child  Routes A & B: outcomes are the same!! In first scenario asked: explain probability with counters Read instruction text Now, I am going to ask you some questions about your beliefs on the child taking care of [his/her] family when the parents grow old. Before that, let’s talk about how I am going to understand your answers better. I will ask you several questions about the chance or likelihood that certain events are going to happen. There are 10 counters. I would like you to choose some counters out of these 10 counters and put them aside to express what you think the likelihood or chance is of a specific event happening. 1 counter represents 1 chance out of 10. If you do not put any counters aside, it means you are sure that the event will NOT happen. As you add counters, it means that you think the likelihood that the event happens increases. For example, if you put 1 or 2 counters, it means you think the event is not likely to happen, but it is still possible. If you pick 5 counters, it means that it is just as likely it happens as it does not happen (50-50, like flipping a coin). If you pick 6 counters, it means the event is slightly more likely to happen than not to happen. If you put 10 counters aside, it means you are SURE the event will happen Question on understanding of the concept about going to the market Pick the number of beans that reflects how likely you think it is that.... A woman in your community will go to the market at least once within the next 2 days. A woman in your community will go to the market at least once within the next 2 weeks. Additional variations 50% of the sample will get female name, 50% male name Order of answer options is randomized Visual aids Used to help caregivers distinguish between situations/scenarios Aids describing parenting styles are labelled based on Situation numbers (i.e. “1” or “2”), and gender type (Male - ”a”; Female - “b”). Situation 1 – Boy/Girl: (Visual aid 1a/b) Situation 2 – Boy/Girl: (Visual aid 2a/b) Aids describing amount of time spent with child are labelled based on route (Route “A”) and time amount type (Low time – “1”; High time – “2”): Low time: Visual aid A1 High time: Visual aid A2  randomization affects whether A1 or A2 is shown first Aids describing child skills are labelled based on route (Route “B”) and skill type (High skill – “2”; Low skill – “1”): High initial skill – Boy/Girl: (Visual aid B2a/b) Low initial skill – Boy/Girl: (Visual aid B1a/b)  randomization affects whether B2 or B1 is shown first Guidelines (I) Ensure proper setup: Asking caregivers set time apart and sit down (for making them note differences between situations) Asking for a table (to do the counters) always remember to pick up counters after use Read slowly and understandable: it is okay to repeat if the respondent asked for it Crucial that the wording is the same for each caregiver (applies to local languages --> more later) Okay to explain a bit more if caregiver does not really understand. Important to limit explanation; we need to strictly follow the scripts and do not provide too much extra explanations It is a hypothetical child so that parents don’t mix it with their own child  it is important to read that well, same for the fact that only described characteristics of the child exist If caregiver asks for additional information, repeat that he/she should base his/her answers on the available characteristics and that he/she should consider all available information We want caregivers to estimate a ballpark number, not the exact value for income Guidelines (II) Neutrality in reading description and transmission of phrasing Particularly for income and when using counters. In the interview, caregiver may first describe how future income will be (e.g., income will be more) or likely child will take care o parent, prompting answer options on scale but saying "like 10" quickly prompts them to choose it Highlighting and pointing at specific differences (visual aids) Each situation, crucial to show and point at pictures to explain together with description Between situations, super important to highlight the difference clearly Ensure neutrality as well and do not convey that a behavior is better than the other Randomization changes (e.g., order of situations) Ensure that the correct visual aid is used when explaining the description Guidelines (III) Find appropriate ways of dealing with caregivers justifying answers and explaining around (e.g., explaining how child should be handled). Reinforce that you don’t judge his/her answers if caregiver justify him/herself Be neutral as well and do not convey that any part of the justification is negative/positive than the other Further questions? LIF instruments (III): Cost scenario Aim: To think about the constraints regarding exerting a certain parenting style Costs: represent a disutility to parents Stress (Limited) time available to parents – our interest for this study Elicit costs using a situation: when administering Intro text; followed by (i) time split between two activities; (ii) situation in which parents decide how to interact with child under different time amounts  Still about hypothetical child and the parents, but a different child Administering Instruments: Cost scenario Cost scenario Ask caregivers for advice for hypothetical parents on how to allocate their time How much time to spend with the child talking or doing something together; Versus How much time to spend doing other things Present a situation: Child of the hypothetical parents has a problem at school Ask caregivers for advice for hypothetical parents about how to interact with child in this situation Two (2) different time amounts available Cost scenario: Intro and things to note Info to enumerators: No addition to make INTRO clearer. Only need to be clear by reading it Otherwise, the experiment is not comparable across you the enumerators Allow to repeat INTRO if something is not clear, but not too much more Info from enumerators to caregivers: Reinforce that there are “No right or wrong answers; and no intention to make judgement” Cost scenario: how it works in practice with the time split Hypothetical parents have [60/20] minutes available: How many minutes should be spent with child talking or doing something together. How much should (hypothetical) parent spend doing other things Hypothetical parents have [less/more] time, [20/60] minutes available: How many minutes should be spent with child talking or doing something together How much should (hypothetical) parent spend doing other things  Variation: In the first time available, 50% (60 minutes) first and 50% (20 minutes) first Cost scenario: how it works in practice when a situation is given Situation: Now, imagine [CHILD B] wants to talk with the parents about an argument with another child at school. [CHILD B] tells the parents that another child has called [him/her] a bad name and [CHILD B] started insulting the other child because of that. The teacher scolded them for this behavior and the situation made [CHILD B] feel upset. Hypothetical parents have [60/only 20] minutes available: How should (hypothetical) parent approach the conversation with child Hypothetical parents have [only 20/60] minutes available: How should (hypothetical) parent approach the conversation with child Cost scenario: how it works in practice when a situation is given Approaching conversation: The parents should tell [CHILD B] to not insult other children. If [he/she] gets into trouble, the parents should punish [CHILD B] by yelling or scolding. The parents should try to understand the perspective of [CHILD B] and discuss what [he/she] could do if the situation happens again. The parents should clarify that if it happens again there will be consequences (e.g., additional housework). Additional variations 50% of the sample will get female name, 50% male name 50% of the sample will get description with High initial skill, 50% get description with Low initial skill Order of answer options for the Situation is randomized Guidelines Neutrality: Especially in reading school encounter situation and answer options Find appropriate ways of dealing with caregivers explaining around Particularly when justifying and explaining around incidence their own child had at school Reinforce that you don’t judge his/her answers Highlighting clearly differences between time amounts when reading school encounter situation Additionally, read clearly the answer options If caregiver says he/she would not advise any of the options, please ask him/her to choose the option closest to his/her advice. Further questions? Administering Instruments: Other modules Other modules Caregiver’s own childhood Ask caregivers about the kind of parenting they were exposed to when they were children Gender bias scale Ask caregivers for opinions on gender bias questions Social pressure Ask caregivers about parenting practices and the opinions of others in their community  Let’s discuss modules from questionnaire