Learning together: A field experiment on parental involvement for children's learning in Uganda

2022-05-09

2022-08-13

The main outcome will be the child's learning (please see below for further details)

Category 1: Child’s learning
The main outcome of interest is children’s learning. We focus on the two dimensions targeted by the interventions, namely math and reading/language, and measure these both through surveys and data from school records.

1.1 Mathematics (two outcomes)
1.1.1 EGMA (index): Share of correct answers (or correct answers per minute for timed subtasks) across 3 categories: Number identification, Number discrimination, and Missing number.
1.1.2 School register data: Ranking within the class in mathematics.

1.2 Language/reading (two outcomes)
1.2.1 EGRA (index): Share of correct answers (or correct answers per minute for timed subtasks) across 11 categories: Orientation to print, phonemic awareness, initial sound identification, letter identification, letter sound identification, oral reading fluency, oral reading comprehension, English vocabulary, letter writing, listening comprehension, sentence dictation.
1.2.2 School register data: Ranking within the class in English and Literacy.

We have four categories of secondary outcomes (please see below for further details): Time use in home learning, parent's attitudes to education, child well-being and household well-being.

Category 2: Time use in home learning
An important source of increased learning is more time spent on learning activities. We measure time spent on learning by the child, and time spent by parents or elder siblings helping the child with its homework.

2.1 Time use child
2.1.1 Minutes spent on homework and learning on a typical day
2.1.2 Days spent on home learning activities in the last week

2.2 Time use others (minutes per weekday spent with target child on reading, homework, and school activities, as reported by the mother)
2.2.1 Mother
2.2.2 Father
2.2.3 Elder sibling

Category 3. Parents’ attitudes to education
The parents’ attitude to education can be key for long-term learning. We measure these attitudes on three dimensions: knowledge, beliefs, and self-efficacy.

3.1 Mother’s knowledge of education
3.1.1 School knowledge: Respondents’ knowledge of P2 school curriculum, a composite index of five questions (Appendix, r10-r13).
3.1.2 Textbook knowledge: Respondents’ knowledge of practice books, a composite index of four questions (Appendix, q1-q5).

3.2 Mother’s beliefs about/attitudes to education
3.2.1 Adequacy of parental belief of child’s performance relative to others. Measured as the distance between the relative performance guess and the actual performance based on the measures described in category 1 (Appendix, H7a-H7d).
3.2.2 Attitudes on the importance of education and its role. Measured as agreement to 4 statements (Appendix, I1a-I1d).

3.3 Mother’s self-efficacy
3.3.1 Parents’ sense of competence scale: We follow the scoring instructions in creating a summary variable for parents’ sense of competence (PSOC).
3.3.2 Parents’ confidence, measured as the average confidence in motivating and supporting the child in learning and at school (Appendix, H10a and h10b).

3.4-3.6: similar as 3.1-3.3, but for father.

Category 4: Child well-being
Increased focus on learning can be a source of both pleasure and stress for the child. For this purpose, we measure children’s mental health and pro-social behavior based on two standard survey instruments: SDQ and ISELA.

4.1 SDQ: We follow the scoring instructions, and create indicators of total difficulties and difficulties in the 5 sub-domains.

4.2 ISELA: Share of correct answers in 3 categories: Stress management, empathy, and solving conflict.

Category 5: Household well-being
We are also interested in whether the learning intervention creates overall benefits to the household or, alternatively, undesirable side effects. For this purpose, we measure happiness and well-being, including whether it has caused tensions and increased violence against children, and whether or not the focus on learning has crowded out time spent on work, and hence led to a reduction in incomes.

5.1 Happiness and Stress
5.1.1 “How happy are you with your life?”, answered on a scale from 0 to 10. We will standardize the answers.
5.1.2 Satisfaction with life: “In your opinion, where are you on the ladder of life at the moment?”, answered on a scale from 1 to 10. We will standardize the answers.
5.1.3 Perceived stress: We use Cohen’s perceived stress scale.

5.2 Violence against children: An indicator for whether the target child experienced any violence during the last 12 months, separated in total, physical and psychical.

5.3 Labor supply and income
5.3.1 Total hours spent by the respondent in income-generating activities during the last month.
5.3.2 Total labor income generated by the household

5.4 and 5.5: Similar to 5.1 and 5.3, but for the father.

Sample selection:

During March and April 2022, we conducted a school survey in the Masaka district. The purpose of the survey was to identify eligible schools, get information about teaching material, and to obtain the enrolment lists of primary 2 (P2) students. We used these lists to sample households to conduct the baseline survey.
Households had to satisfy the following eligibility criteria: One of the children was enrolled in P2 and continued to be so in term 2; the child was between 6 and 12 years old; the household did not intend to move away during term 2; and at least one adult caregiver was present in the household.
In total, we identified 27 schools, with 7 to 87 eligible P2 students. The total sample of eligible households (and thus students) is 933.

Data collection:

We conduct three rounds of surveys:
1. Baseline survey: May 2022 - before the start of the second school semester.
2. Short-run follow-up survey: August 2022 - at the end of the second term. We collect data on the same set of indicators as at baseline, including the childhood development survey. Moreover, when available, we survey the father and the oldest sibling (living in the household).
3. Endline survey: December 2022 – the end of the third term. We collect data on the same set of indicators as at baseline and the short-term follow-up survey, including the childhood development survey.

In addition, we collect school administrative data on the children (grades and attendance) for the entire school year.

Randomization was done in office by a computer

individual

No

933 households

933 households

461 households in treatment, 472 households in control