Sample Surveys Lecture 2

SAMPLE SURVEYS *.V.K.JAYARAGHAVENDRA RAO

WHAT IS A SURVEY? A survey is an investigation of the behaviour/opinions of a Group of people or a sample. A sample is one part of a whole that can be examined in order to know what the rest is like Eg., technology survey on marginal farmers. Research questions appropriate for a survey? Not all research questions are appropriate for a survey, you may have a botanical survey, which is entirely different, while in sample surveys we deal with the behavioural aspects of human beings viz., •Behavior •Attitude/beliefs/opinions •Characteristics •Expectations •Self classification •knowledge Before we go into Sample survey in detail one should know the difference between an experiment and survey Experiment: an experiment usually means a lab or a field experiment done mostly on other than human beings , in some cases even human beings may be the object of an experiment as in case of medical research, it should have the following elements in order to qualify for an experiment. 1.subjects respond to treatment created by the researcher 2.Causality is shown by timing of treatments 3.Associations between treatments are observed 4.Controlling alternative explanations 5.Physical control on treatments Survey: usually refers to collection, classification and inferencing on the human behaviour and the predispositions towards a psychological object thereof also referred to as the attitudinal aspect of a phenomena or technology in question for agricultural researchers, it should have the following elements in order to qualify for a sample survey 1.Sample many respondents and ask the same question 2.Measure many variables 3.Test many hypothesis 4.Infer temporal order from questions about past behavior 5.Survey researchers measure variables, which represent alternative explanations 6.Survey research is correlational 7.Approximations of the rigorous test are done

NAARM LECTURE NOTE-2001- Sample surveys

Types of surveys: There are many types of surveys popular among them are: 1.Mail and self-administered questionnaires :used by social scientists 2.Telephone,email,computer-based interviews/surveys: used by market researchers 3.Face to face interviews: used by market survey researchers , social scientists etc., 4.Panel technique-select group and study attitude(opinion poll)- used by psephologists- study of political issues and trends 5.PRA and special situations: for need identification of rural folks Steps in sample surveys 1.Design and planning phase: this is the most important phase of sample surveys and it shall be done in detail because once a plan is finalized for a sample survey then the rest will be only execution 2.Data collection phase: in this phase the data will be collected as per the scheme finalized during the design and planning phase 3.Data analysis and presentation: once the data is collected it shall be analysed using any of the social science statistical packages like the SPSS-10.1 version currently available 1.Design and planning phase •Objectives of the study: spell out the objectives of the study usually in verbal form, like to identify, to analyse, to categorise, indicating what you will do with the survey. •Decide on type of survey, respondent and population •Develop the survey instrument/questionnaire/semi structured interview schedule(PRA- take the help of judges/experts to do this Write questions to measure variables Decide on response categories Organise question sequence Design questionnaire layout •Plan a system for recording answers, items of information: that is go through the questionnaire/schedule and plan who will interview and who will record •Pilot test the instrument and train the interviewers: test the instrument/questionnaire/schedule in an area other than where the actual survey is to take place

Sample surveys 2

•Sampling: Draw the sample by design or sampling design: the following has to be considered before drawing the sample Define target population: all the paddy growers who have grown paddy for the last 5 years, on an area of more than 5 acres under tank fed irrigation method Decide on type of sample : let us say stratified random sample Develop sampling frame: 10 farmers from each of the taluks from the eight districts, with 10 taluks in each district, so the sample size comes to 800 Decide on sample size: say n=800 Select a sample: after all this finally select the sample taking the help of revenue officials or extension workers etc., DIAGRAMMATIC REPRESENTATION OF DRAWING A SAMPLE Sampling

Universe

draw

Sample which represents the population

Why should we sample for a survey? Normally people ask this question as to why one should sample for a survey, why not any one on the roadside or under the banyan tree in a village, the importance of sample in a survey is to : • make sample more efficient •Manageable •Cost effective-Provide for lowest possible Cost Terms used in sampling surveys Population: a large pool of cases or elements(all the cases/elements)/even the term universe is used Target population: specific pool of cases with a criteria Sampling frame ; a specific list of all cases matching the criteria Sample surveys 3

Sampling element: unit of analysis or case Sampling ratio: is the ratio of the sample to the population Sampling design= Rule to draw the sample Random sample= is not a haphazard sample but based On the laws of probability How large should your sample size be? This is often the question the researchers will ask because a large sample size may be good from the statistics point of view, but the researcher will have problems with such a situation, however the following guidelines help the researchers draw samples of appropriate sizes: •Smaller the population bigger the sample size and vice versa •1000 popn-30 percent should be the sample •10,000 population 10 percent should be the sample •For large population 1,50,000 –1 percent sample •For very large population(10 million and above)-0.025 percent Principle of sampling : the principle of sampling is to ultimately reduce the errors due to sampling , increase the variance due to the variable in question or it is also called the true variance, it is represented by a small equation below: V=Vt +Ve Where V= is the total variance, Vt=is the true variance due to the variable in question and Ve= is the error due to extraneous influences or the sampling errors in our case. if the sample is small –small increase produce big gains in reducing the errors, eg., when you increase the sample size from : 50 -100-errors reduce to 2.1% from 7.1% 1000 -2000-errors reduce from 1.6-1.1% Sample size depends upon 1.Degree of accuracy 2.Degree of variability 3.Number of variables examined When you wish to improve accuracy, variability, and the number of variables increase sample size Types of samples

Sample surveys 4

Samples are of two types the non-probability sample and the probability sample, the former does not use the sampling frame, while the latter uses it, the following table depicts the type of samples under these categories with their description: PROBABILITY SAMPLE

NON-PROBABILITY SAMPLE

)

)

(sampling frame

(no sampling frame Haphazard: select anyone convenient

Quota: select anyone in predetermined groups

Purposive ; select anyone in hard to find target population

Snowball: select people connected to one another

Simple: true random procedure

Systematic ; select every kth person

Stratified ; randomly select people in predetermined groups

Cluster : take multistage random sample in several levels

NON PROBABILITY SAMPLE A non-probability sample is one, which makes no assumption about the probability distribution of the population a few examples are given below Haphazard sample : it is a sample of convenience, and should not be confused with random sample, the characteristics of this sample are: • select anyone convenient •Seriously misrepresents the population When to use: It is normally used for entertainment value(eg., MTV Bakra programme is a haphazard sample )

Sample surveys 5

• •

Shortfall: the shortfalls of this sample are Gives a distorted view Seriously misrepresent the population

Quota sample : This is also another type of non-probability sample the steps in drawing such a sample are: • identify categories (male/female), the quota of sample size has to be filled in these categories . •5 males 5 females under 30 years of age •Then use haphazard sampling When to use: • For entertainment value(MTV Bakra) Shortfall: • Distorted view • Seriously misrepresent the population Purposive sample : in this type of sample the researcher uses a particular sample purposively eg., all PhDs ‘s in Agronomy in an ARS discipline of a particular year for study purpose, the characteristics of this sample are : • they are judgemental •Uses the judgement of an expert •Informative cases When to use: • When difficult to reach • To identify a case in depth Shortfall: • Cannot generalize for a population • Purely opinionistic Snowball sample : snowball sample comes from the name a small piece of ice while rolling becomes a big snowball in the sense picks up the snow( which is related) , that means sampling people who are related to each other with a criteria, community groups, clique groups, friends circle etc., the characteristics are: •-interconnected networks •sociometry •Key informants When to use: • Used in PRA and RRA

Sample surveys 6

Shortfall: • Cannot generalize for a population • Purely opinionistic • Has temporal and spatial limitations PROBABILITY SAMPLE: A probability sample is one, which makes assumptions about the probability distribution of the population, and the sample represents the population in terms of the statistic a few examples are given below: Random sample : • use random number/table •With replacement •Without replacement When to use: • When population is well defined • When sampling frame is established Shortfall: • Cannot generalize for various groups Systematic sample : a systematic sample is after drawing the sample from a lot(population) number them and select the kth elements • number the sample •Select the kth elements When to use: • When population is well defined • When sampling frame is established Shortfall: • Cannot generalize for various groups Stratified sample : a stratified sample is to divide the population into various layers or strata and then draw random samples from each of the strata so defined , eg., stratify the states/districts/taluks/villages/and then farmers growing paddy, it uses both the criteria and representation of elements in the sample ,so that meaningful conclusions and inferences can be drawn, the steps to draw such a sample are • divide the populations into subpopulations •each subpopulation is called a strata •Random sample from sub population or strata using random or systematic sampling Sample surveys 7

When to use: • When variation is there in the strata • When population is well defined • When sampling frame is established Shortfall: • Time consuming • costly Cluster sample : it is to be used when the population list is not there and the population is spread over a large area, that means when spatial distribution is large, eg.,to study the attitude of paddy growers towards fertilizer in India, then randomly select the states /districts/taluks/villages /then farmers whose cluster list is there means list of small/marginal and large farmers, select the farmers from the list of this cluster at random. The key points of using such a sample are: • when the population is dispersed spatially on a large area •Randomly select districts then taluks then villages(3 stage)then farmers •When clusters list is available When to use: • When population list is not there • But the list of clusters(subgroups are there) • When sampling frame is established Shortfall: • Time consuming • Requires lot of information Random digit dialing sample : Nowadays when the telephone directories are available on the internet/CD’s/books the market survey researchers normally use this sample to find out the attitude of the consumers towards various products, Agricultural researchers can also use this on farmers as the telephones are there in nook and corner of the country and the names may be found from rural listings also. The characteristics of this sample are : • you can stratify/cluster/random sample also •Uses the telephone directory •Population is the telephone directory When to use: • used for telephone interviews • When sampling frame is established Shortfall: • Not so much used in India and in agricultural situations Sample surveys 8

•

May suit only big farmers

Why sample survey predictions are wrong? Some examples for thought?? • 1936-F.D Roosevelt won by a landslide but Alf landon was Predicted-mistakes in sampling(sampling frame was wrong) • It excluded people without telephones or automobiles-lower Income favoured Roosevelt • 1948-gallup predicted the wrong candidate • 1996-exit polls predicted the wrong candidate/party After the sampling and the sampling frame is established, the questionnaire/schedule /and other instrument of data collection is developed the next step is to go ahead for data collection which is the second phase in a sample survey 2.Data collection phase: steps in organizing the data collection •Organize the survey •Locate and contact the respondents •Make introductory statements or provide instructions •Ask questions and record answers •Thank respondent and continue to next respondent •End data collection and organize data After the data collection is over the data needs to tabulated and analysed to draw inferences and conclusions 3.Data analysis and presentation: steps in data analysis and presentation •Tabulate properly using MS excel, any spreadsheet •Use appropriate statistical tools •Proper interpretation using your subject and experience Some of the advantages of sample surveys are : Advantages 1.Great deal of information can be collected 2.Large population can be used 3.Reduces labour 4.Information is accurate 5.Variety of information can be collected Some of the disadvantages of sample surveys are : Sample surveys 9

Disadvantages 1.In-depth study not possible 2.Costly 3.Time consuming 4.Scope for sampling error questions of Dr.V.K.Jayaraghavendra Rao Answer the following questions 1. Research questions appropriate for a survey?(2 marks) 2. What is a survey? (2 marks) 3. Differentiate between survey and experiment?(3 Marks) 4. What are the differences between quota sampling and random sampling?(2 marks) 5. What are the advantages and disadvantages of sample surveys(3 marks) QUESTIONNAIRE FORMAT FOR SAMPLE SURVEYS Year 2000 Problem 1. How familiar are you with the Y2K problem? (check only one) [ ] Very familiar (you know what has caused it and it's potential affects) [ ] Somewhat familiar (you've read a thing or two and have a general idea of what's going on) [ ] Neither familiar nor unfamiliar [ ] Somewhat unfamiliar (you've heard about it, but that's all) [ ] Very unfamiliar (you know nothing about it) 2. When thinking of the potential affects of Y2K, are you: (check only one) [ ] Very optimistic (everything will be fine) [ ] somewhat optimistic (things might go a little crazy, but generally,everything will be fine) [ ] neutral [ ] Somewhat pessimistic (we're looking at tough times ahead) [ ] Very pessimistic (no one gets it, things are bad!) 3. Would you describe your primary banking relationship as being with (please check one): [ ] A large national/international bank [ ] A regional bank [ ] A local/single city bank [ ] Don't currently have a banking relationship 4. With your bank(s), do you have (check all that apply) [ ] A personal account [ ] A business account

Sample surveys 10

5. Do you believe Y2K will directly affect your bank(s)? (check only one) [ ] Yes [ ] No [ ] Don't know 6. If yes to Q5 above, do you believe Y2K-relating affects will be: (check only one) [ ] Short-term (a few days to a few weeks) [ ] Moderate-term (a few weeks to six months) [ ] Long-term (six months to a year) [ ] Don't know 7. If yes to Q5 above, do you think Y2K will affect: (check all that apply) [ ] Your banks administrative functions (loss of records, communications problems, etc.) [ ] The security of your money on deposit [ ] Information regarding loans, credit cards and other depts [ ] None of the above 8. If you feel your money may not be secure in the advent of Y2K, which of the following statements describe your thoughts/actions: (check all that apply) [ ] The FDIC will cover my accounts against Y2K-related loss [ ] I will compile complete paper records of all accounts prior to Jan 1, 2000 [ ] I will withdraw extra cash prior to January 1, 2000 in preparation for possible problems [ ] I will withdraw most or all funds on deposit in preparation for possible Y2K-related problems [ ] None of the above 9. Have you made or would you make any changes in banking relationships based on potential Y2K problems? [ ] Yes, I have moved my account to a bank with solid Y2K compliance [ ] Yes, I would move my account if I found a bank with solid Y2K compliance [ ] No, I won't move my account in Y2K preparation 10. Is your bank providing information on its Y2K preparedness? [ ] Yes [ ] No 11. Would you describe your bank(s)' communication with you about its level of Y2K Sample surveys 11

preparedness to be: [ ] Satisfactory [ ] Unsatisfactory 12. Do you believe the banking industry is being forthcoming about its level of preparedness for the year 2000-date change and the effect of the date change on the financial services industry? [ ] Yes [ ] No 13. Which of the following have you considered regarding potential Y2K problems: (check all that apply) [ ] availability of funds on deposit with banks, savings and loans, and securities firms [ ] availability of 401K, IRA and other retirement accounts [ ] use of credit cards (they might not work, the companies might lose information about you, your account might vanish, etc.) [ ] use of personal computers (they might not work) [ ] use of personal care and appliances items that have microprocessors in their designs [ ] travel safety (traffic lights might not work, airline instrumentation and tracking devices might not work, etc.) [ ] utility availability (power, water, telephone, etc.) [ ] Internet use 14. To-date, have you done anything to prepare for Y2K? (check all that apply) [ ] I've debugged (or at least started debugging) my personal computer [ ] I've started asking banks, utility providers, etc. for Y2K preparedness statements [ ] I've moved accounts, closed accounts, etc. in preparing for Y2K [ ] when shopping (for computers/electronics/etc), I look for Y2K compliance statements [ ] I've taken other measures in preparing [ ] I've done nothing to prepare Thank you for your participation.

Sample surveys 12