RESEARCH: MID-DAY MEALS

[Introduction] [Survey Questionnaires] [Survey Guidelines] [Findings]

Survey Guidelines

1. General
   • Background: The main purpose of this survey is to evaluate the Mid-day Meal scheme run by the government in Government and government aided schools. It is also an opportunity to learn about the economic and social circumstances of destitute households. This is not just an "academic" exercise; we are hoping that the results will help to make a case for a program with greater scope and better quality.
   • Ethical aspects: It is important to conduct the interviews in an ethical manner, and in particular to avoid taking advantage of the position of power in which you may find yourself. For instance, make sure to introduce yourself, to explain the purpose of the survey, and to be considerate in accepting any refreshments that may be offered. Also, do not press the respondent for answers, when he or she is reluctant.
   • Quality of the data: The success of this project depends crucially on the quality of the data. Your most important responsibility as investigators is to ensure that the data are as reliable and accurate as possible.
   • Personal observations: We would like to associate you not only with the field survey but also with some of the discussions that will lead to the report. This means that we don't simply expect you to fill the questionnaires mechanically; we also look forward to learning from your personal observations and experiences. Even in the questionnaires, space has been provided for "personal observations"; please make liberal use of this opportunity to share your observations.
   • What to do in a crisis: If there is a problem, you should call the Survey Coordinator and take action in consultation with him/her. Make sure not to interrupt the fieldwork, unless advised to do so by the Survey Coordinator.
2. The Questionnaires
   There are six questionnaires:
   1. State questionnaire
   2. District questionnaire
   3. Village Education Committee questionnaire
   4. Headmaster questionnaire
   5. Cooks Questionnaire
   6. Parents questionnaire
   They should be filled roughly in that order. In any case, you should start with the "Village Education Committee questionnaire" and "Headmaster questionnaire", and then go on to the questionnaires for the Parents and cooks.
3. Basic principles of interview
   • Normally, two investigators should work together on each interview. One investigator should conduct the interview; the other should record the answers on the relevant questionnaires.
   • Introducing yourself: The first step is to introduce yourself and to explain the purpose of the survey. You should also seek the respondent's consent for the interview. If he or she is unwilling, do not insist.
   • If at all possible, avoid visiting households with the sarpanch or other local netas. If they follow you around, try to explain to them that you have strict instructions to visit the households without escort. There is no harm in getting help (e.g. for translation and related purposes) from local residents who don't have too much of a vested interest in this matter, e.g. a local teacher or social worker.
   • If the respondent is a woman, try your best to get the answers directly from her rather than from male relatives, neighbors, etc. In such cases, it may be helpful to be accompanied by a local woman, e.g. the anganwadi worker or a member of the local mahila mandal.
   • Probing: Some questions may not immediately elicit a response. In such cases the investigator will have to "probe" a bit. By this is meant that a few extra questions will have to be asked to make the question clearer to the respondent, or to get a more precise answer from the respondent. However the investigator must make sure that the meaning/sense of the question is not lost/altered in the process of probing.
   • Try to avoid "leading questions", i.e. asking a question in a manner that encourages the respondent to answer in a particular way. We have tried to avoid leading questions in the questionnaires, but you will have to be aware of this potential danger also. Example: it is better to ask, "Do you feel that the Mid-day Meal Scheme is helpful or unhelpful?" instead of "Do you feel that the Mid-day Meal Scheme is helpful?"
4. Filling the questionnaire
   • Clarity is paramount. In particular, make sure that your handwriting is legible, and that all numerals are clear. Illegible information is useless.
   • As far as possible, please do not leave any 'blanks' in the questionnaire (blanks tend to create problems at the "coding" stage). Instead, use a dash ('-') for 'not applicable' and 'NR' for 'no response'. The distinction between the two can be very important, please make sure to take it into account.
   • If you find that an answer does not "fit" into the pre-specified format, you can record the answer in "long hand", as long as the information is clear and complete. Example: Suppose that in response to the question "Are you satisfied with the scheme? The only two responses are satisfied and dissatisfied and the actual response is that they are satisfied to the extent that the quality of food is good but dissatisfied to the extent that the quantity is too little. In such a case it would not be right to just tick on one of the two choices, it would be advisable to take a moment to write down such a response.
   • In places where there is space for "qualitative information", please make an effort to note any interesting details that may be relevant to this study (instead of "doing the minimum"). If you run out of space to note any details, please write at the back of the same page.
   • Try to avoid "bogus answers". If a response strikes you as non-credible, it's better to write "NR" than to record a bogus answer.
   • Use the relevant spaces to record the answer, especially the quantitative answers (there is a clearly demarcated "box" for each numerical answer). Try to avoid writing in margins, but if you do, use the left-hand margin, not the right-hand margin (using the right-hand margin may confuse the data-entry operator).
   • Please make sure that all questionnaires are safely stapled. Loose pages are a headache since one may not know which questionnaire they belong to.
   • Use Arabic numerals (1,2, etc.), not Roman (I, II,...), Hindi, etc.
   • In case of questions that have multiple answer choices let the respondent answer the question on his/her own and you must record depending on the relevant answer choices. If you ask the different choices directly you may end up leading the respondent. Sometimes when the respondent does not understand the question then the answer choices may be used as cues.
   • In cases where you have to choose between different options, and each option has a "code" attached to it, please circle the relevant option, including the "code".

   Example: Q 15 from Part 6: If dissatisfied, what do you feel are the main reasons for low quality of the food? (Tick more than one response if necessary)
   (1) Inappropriate menu
   (2) Grain is of low quality
   (3) Ingredients are of low quality
   (4) Absence of a competent cook at school
   (5) The cook is careless
   (6) Other reasons (please specify)
   And the responses are the first two then make sure to circle the relevant codes as well (1) and (2) in this case.
   Also remember to record these responses in the extreme right column.
   

5. Question-wise clarifications and comments
   • Part 4: Headmaster Questionnaire
     • While questioning the Headmaster it is important for you to understand the difference between enrollment and attendance. Enrollment is the number of students that have enrolled for admission to the school, i.e., the students whose names appear on the attendance register. Attendance on the other hand is the number of students actually attending school, i.e., present on a particular day. Data on enrollment and attendance figures is crucial for the purpose of our survey and you must take utmost care while recording these figures. Getting figures for enrollment would be fairly simple, this you can get from the school attendance/enrollment register for the year 2001(last year) and 2002 (this year). Getting figures for attendance would be a more challenging job. We want you to record the total attendance for each day of the month preceding the survey, for all students of class I and for classes I-V taken together. We also want the same figures for the SC/ST children in class I and for classes I-V taken together. For Sundays and other holidays put a cross "X" in the relevant cells and for days when the data is missing put "M" in the relevant cells. And please fill all the cells and blanks that need to be filled out (empty cells can make the data unusable for us).
     • A "double meeting" system is one in which attendance is taken twice, once in the morning and once after lunch break. In these schools the attendance figure may appear conspicuously magnified. If the school you visit has a double meeting system, just record the figures as they are and mention that they are double meeting figures.
     • Qn 7, here we want to know the number of classrooms only.
     • Qn 34 and 36, if the headmaster in unable to calculate the percentage figures you try to do it using the figures of last year's enrollment (LE) in the given month and the current year's enrollment (CE) in the same month by using the following formula: (CE-LE x 100)/LE
     • Qn. 37 and 38, please read the question carefully, here we are talking about absenteeism and not the attendance. Make sure to explain the difference between the two. The question 37 reads " In your perception, has there been any change in pupil absenteeism this year compared with last year?" If the response is yes, then you go on to the next question which concerns the direction of change. The given coded answers are "increased", "decreased" and "no response". If the headmaster says that it has increased then the implied meaning is that attendance has gone down, so double check with him/her to make sure they have understood the question properly.
     • Qn 44, this question reads, "Have some upper-caste parents ever objected to their children sharing a meal with children of other castes?" If the response for this question is "No" then Qn 45 and Qn 46 (parts 1,2 ,3 & 4) are not applicable, so you must put a dash for each of these questions.
     • Qn 49, this question is about the water arrangement if
       • the response is that water is brought to school from the nearest source (code 2)then Qn 50(part1) becomes not applicable as that asks about the source of water within the school itself. So put a dash for Qn50.1 and go on to Qn50.2.
       • the response is that there is a source of drinking water at the school itself(code1), then Qn 50(part1) is applicable but Qn 50(part2) becomes not applicable.
       • the response is that children go to the nearest source to drink water, then Qn 50 part 1 and 2 both are not applicable, so put dashes for them and go to Qn 50 part 3.
   • Part 5: Cooks Questionnaire
     • Qn 14 and Qn 15, through these questions we want to know how the fuel is arranged for the MDM. The response to Qn 14 will be "yes" only if fuel as such is provided to the cook. If the cook is given money for the fuel then the response to Qn 14 would be "No" and the payment for fuel should be recorded in Qn 15.
     • Qn 17 try to elicit as much as you can regarding the MDM. Probe into how the cook actually cooks the food starting from cleaning the grain to giving the cooked meal to the school.
     • Qn 30 from this response try to gather who and what may have been the main reasons for his/her selection and then verify in later questions.
     • Qn 33, this is about the cook's remuneration. Please note that this can be in monetary figure or in kind. Whatever it is please note the monthly remuneration.
   • Part 6: Parents Questionnaire
     • Sampling: You should select the parents according to the following criterion: We will be visiting three villages in a particular block. Take parents of class I children in the first village, of class III children in the second village and class V children in the third village of each block. Take the attendance register of the relevant class and take the first SC/ST girl and then the first General girl, then the first SC/ST boy and then the first General boy. (follow this in all the blocks). This way you will have a sample of four parents in each village.
     • Qn 7 if the respondent does not know the amount in acres then write in whatever unit they know and get the local conversion into acres.
     • Qn 8 this question is based on your observation of the village as a whole. So be observant and record what you feel is appropriate.
     • Qn 10,15,25 while asking this question observe how they have been phrased. We have tried not to make it a leading question, so while asking make sure you preserve the unbiasedness of the question.