How Many Samples Could I Draw?

Sample selection

How to draw a sample

Well-nigh of the methods described in this transmission arc qualitative, that is, they depict without necessarily quantifying observations. There might be times, however, when more quantitative data is required. Perchance you suspect that your observations, if systematically recorded and analyzed, will reveal patterns or associations which are not obvious without special testing. This involves the use of statistical analysis, a subject outside the scope of this manual; at that place are plenty of good books on the subject as well as powerful and easy-to-use computer applications.

In short, statistical tests allow y'all to draw informative conclusions about very big groups of people, things, or occurrences past observing a small representative sample of these people, things, or occurrences. Since this manual is about collecting information, information technology is worthwhile knowing what constitutes a representative sample.

Imagine that the population you lot are studying is a pot of stew with all sorts of vegetables, beans, and meat. Taking a spoonful from the height might lead one to believe that the stew is thin. A spoonful scraped from the bottom of the pot might suggest an peculiarly hearty stew. To learn the true nature of a stew, you must fires stir, so sample. In terms of your population under study, this means y'all will need information which adequately represents the diversity of the population. It does non mean you have to stir people in a big pot! Only it does mean you have to know something about the population before yous start.

one Decide Your Population. What practice you want to study? In what area? For example, yous might decide to report the indigenous cropping practices in the Western Highlands in your country. Your population might therefore be crop raisers in that region.

2 Decide on the unit of measurement of study. Are yous interested in individuals, couples, households, families, organizations, or villager? In plants, animals, herds, fields, farms, or watersheds? (For the crop multifariousness report, the unit of study might be households—though you demand to decide who in the household—male person or female, old or immature—you should interview.)

iii Determine Your unit of measure. For the study on ingather varieties, this could be the yield per hectare (or the local unit of measurement of expanse) of each variety. It is likely that yous'll be interested in other unite of measure, as well. For instance, you might also desire to measure out the number of varieties grown, the level of organic and inorganic fertilizer used, the fourth dimension of planting and harvesting, the number of days labor put into a unit area, and other characteristics.

iv Work out how large the sample size will have to be. The sample must be big enough, otherwise your data, no matter how carefully nerveless, and your statistical analysis, no matter how skillfully

performed, will be useless.

The minimum sample size depends on several things:

- How much time and money practise yous have?

- How large is the population that you want to generalize?

- How much variability is in that location in the population?

- How many subgroups do you lot want to separate the population into during your analysis?

- How confident exercise you want to be that your findings are correct? (This is measured by the confidence level: a conviction level of 90% means that y'all can exist 90% confident that the true value of a mean will be reasonably accurate. Gee a statistics volume for details.)

The following table shows the minimum sample sizes required for various population sizes with a 90% confidence levels.

Population size (Due north)	Minimum sample size (northward)	% of population
fifty	33	66%
100	l	fifty%
200	67	33%
500	83	17%
1,000	91	9%
2,000	95	4.8%
5,000	98	ii%
ten,000	99	i%
50,000	100	0.ii%

Notation that these are minimum numbers required. If-you dissever the sample into subgroups for assay, or if y'all demand a college level of confidence, yous volition demand to sample more than this. For a 95% confidence level, you lot demand to sample almost four times the number due north shown in the table. In general, for populations of more 5000, a sample size of almost 100 is enough for nigh simple statistical analyses at ninety% level of confidence (400 for 95% confidence level). You should not use samples of less than about 30 for statistical analysis.

Note

Draw a bigger sample than yous need

It is usually a good ides to draw a bigger sample than yous actually demand (say, 15 to 20% larger than y'all need). You need to exercise this considering some people may not be in when y'all visit, may refuse to answer your questions, or requite responses that you decide yous cannot trues. If you start off with a bigger sample than y'all need for the statistical analytic, you can always replace these people with the extra respondents.

In many instances, you might be tempted to go to the house next door if someone is not at abode when you visit for an interview. You can do this as long every bit you are consistent. Beware of introducing bias—for instance, by interviewing a asymmetric number of elderly people, since these are the ones who arc at home when you call.

5 Brand a listing of all members of me population. This is your sampling frame. You tin can make such a list in several ways:

- Ask local people to brand such a list. Make certain that they include "invisibles" such as low-degree families or landless people.

- Develop the list through one of the other techniques described in this manual, such equally Mapping or Sorting and ranking.

- Obtain a list from local authorities. Possibilities include the village leader, local land records, lists of schoolhouse children (from the local schoolhouse), patients (from the clinic), or cooperative members (from the coop). Beware: these lists might be biased to one department of the community. But they tin can be a adept ground on which to build a more than consummate list with the help of local people.

Making a reasonably complete list can be very difficult, especially for large populations. If y'all cannot do 50, don't worry; you can use other methods of sampling (described below).

6 Describe the sample to report. You can apply i of the techniques described below.

Drawing lots

1 Write all the names of the people (households, farms) in your sampling frame on small pieces of paper.

two Fold the papers uniformly, put them in a big pot and mix them well. (For a stratified sample [encounter beneath], put the names for each subgroup in a unlike pot).

3 Then reach in the pot, without looking at the names, and pull out at random the number of pieces of paper (n) you lot need.

4 It is a adept idea to pull out extra pieces of newspaper (extra households) to serve as substitutes in case some households take moved or cannot be located or contacted afterward several attempts. Afterwards two or three follow-upe without success, yous can use the substitutions. These should as well exist chosen randomly.

Cards

This method tin be used for drawing samples where there are less than 52 (the number in a deck of carafe) individuals in the sampling frame.

1 On a slice of newspaper, list the names of all individuals in your sampling frame.

2 Number the list (1, two, three, etc.).

3 Shuffle a deck of ordinary playing carafe. and draw out northward cards at random from the deck.

4 Write down the figures that correspond to the carafe you have fatigued, according to the following table.

Cards

5 Select the names from the list that correspond to the numbers you have written downward.

Calculator

Some calculators can generate random numbers at the push of a button. These numbers arc typically a decimal between 0 and one. To utilize the numbers generated, you tin either:

- Multiply each number by the number in the population, plus i. For instance, if at that place are 500 people in the population, multiply each of the random numbers generated past 501. (The extra 1 is necessary to brand sure that the 500th individual also has a chance of existence selected, since the random numbers generated range from 0.000 to 0.999, not 1.000).

- Alternatively, you can ignore the decimal betoken in the random numbers, and ignore any numbers that are higher than the total of your population.

Random number table

1 Make a list of all individuals in your sampling frame.

two Number the [fat (1, two, 3, etc.).

three Determine how many individuals y'all need to sample. This is your sample size, n.

4 In the tabular array of random numbers on page 35, locate a starting signal past closing your optics and pointing with a pencil to whatever position on the page.

five Starting at the number you have but chosen, read downward vertically and choose the next north numbers in the cavalcade. If you need fewer than 100 individuals, accept the start two digits in the cavalcade (or the middle 2, or the last two: information technology doesn't matter, as long as yous're consistent). When you become to the bottom of the column, go to the height of the next cavalcade.

6 Select the names from the list that correspond to the numbers yous nave called from the table.

Systematic random sample

This method is easier than random sampling if your list contains a large number of individuals (say, more 500).

1 Obtain a fiat of all individuals in your sampling frame. Count how many in that location are (say, there are 15,000 people).

2 Decide how many individuals you want to sample (say, you want to sample 200 people).

3 Split up the number of individuals by the number in the sample (i.e., xv,000/200 = 75). This is the sampling interval.

4 Select i individual from the list at random every bit a starting bespeak.

5 Select every subsequent 75th person (or whatsoever your sampling interval is) from the list.

If yous cannot go a listing of the members of the population

You tin still generate a sample that will exist approximately random. Here are some suggestions.

Using a map

i Obtain a map of the area (or ask local people to make one—run across Mapping)

ii Select locations at random from the map. For example, you lot can draw a filigree of squares over the map, number the squares, and select squares at random using 1 of the techniques described in this section. You can then go to the locations called and select the nearest house (field, farm) to the chosen location.

Random road

1 Select a location at random.

two Place further addresses (fields, farms) by taking alternate left and righthand turns at road junctions and calling at every nth accost (field, farm, etc.) en route. (The number n depends on the sampling interval, see in a higher place under Systematic random �ample.) Alternate between the right and left sides of your path.

Using a transect

1 Identify a series of transects at random (possibly using a map, ace to a higher place).

ii Walk along each transect and select every nth house (field, subcontract). Alternating between the right and left sides of your path.

Cluster sampling

Perhaps yous want to interview a sample of people who live in villages scattered over a large surface area. At that place is no listing of all the people in the area, and it would be impossible to brand one. This is where cluster sampling tin aid.

1 Brand a list of all the villages in the area (these are the "clusters").

2 Determine how many villages you tin beget to visit (this may depend on coin or time available). In full general, select every bit many equally possible.

three Select this number of villages at random.

4 Within the villages y'all have selected, cull people at random using ane of the other methods described in this department.

5 It may be necessary to draw a multistage cluster sample in some instances. For example, you could select several districts at random, and so cull villages at random inside the districts, and cull people at random within each village.

Snowball sampling

Snowball sampling does not produce a sample that can be analyzed using standard statistical tools. Only it is a useful way of identifying hard-to-find individuals. See Identifying indigenous specialists for a description of how to use this sampling method.

Stratification

Your population might incorporate an important subpopulation that might exist underrepresented in a elementary random sample. For case, you lot might be interested in the indigenous cognition of large landowners compared to smaller-scale farmers Merely if in that location are merely xx big landowners with more than two hectares of land and 5000 Small-scale farmers with leas than 2 hectares, you would need to sample a very large number of people at random in order to be reasonably sure of interviewing enough large landowners.

The answer to this problem is chosen stratification. Y'all first split up the population into subsets, or strata. You tin get local people to help you do this (ace Identifying indigenous specialists and Sorting and ranking). So randomly select individuals from within each subset. for instance, you might select 100 households with less than 2 hectares and all 20 of those with more than two hectares.

Compiled by Ricky Grand. Anunciacion, David Abbess and Paul Mundy Sources: Krejcie and Morgan, 1970, (Bernard, 1988, Babbie 1986

Table of random numbers

This table contains yard 4-digit random numbers. See How to describe a sample for how to use the tabular array to select a sample at random.

Table of random numbers

Identifying indigenous specialists

Definition

A method employing breezy questioning and diagramming to place individuals with specific know-how.

Purpose

To identify indigenous specialists. Indigenous specialists are customs members who have special skills or expertise in one or more subject areas or who practice a profession (e.g., healers). The method can exist adjusted to identify other types of individuals—such every bit conclusion makers, innovators, political opinion leaders, etc.

Materials Notebook

- Pen

- Manila paper

- Marking pen

Possible Approach

1 Define the topic you want to investigate (such as farming or health). Be as clear as possible about its focus and scope.

two Identify the type of people who can help. Information technology might be useful to beginning with people who are involved in activities relating to the topic. For instance, if the topic is farming, you should inquire people who do subcontract work (both men and women). If the topic is cooking, ask family members who do the cooking.

iii Select a sample of upwardly to 20 such people. The number of people will depend on the topic. For highly specialized topics (such as irrigation tunnel edifice), you will probably need only a pocket-size number of people in the initial sample, since only a few people are probable to exist knowledgeable nearly these subjects.

iv Enquire each person to name the people in the village who know the moat about the topic. Ask each respondent to proper noun up to four people.

five Write downwardly the names of these people and where y'all can discover them.

6 Visit each person named. Enquire them to name the people who they call back know the about about the topic. Add the new names to the chart and visit these new people.

vii If necessary, repeat steps 4 through 6 until no new people are named.

8 Draw a diagram showing all the people named. Draw each person as a circle with the name underneath.

nine Describe arrows from each circle pointing to the circles of the individuals each has named. Count and tape the number of arrows pointing toward each circumvolve.

10 The individuals alluring the highest number of arrows are the indigenous specialists for that topic.

Value

- This method quickly generates a list of individuals with specific skills or characteristics.

- These individuals can supply valuable information about their detail area of expertise. (See other methods in this manual which rely on indigenous specialists or fundamental informants.)

Dos and don'ts

- Do repeat the process for other topics as required. A specialist one topic (such every bit farming) is not necessarily the most knowledgeable person on another subject area (such as cooking).

- Don't rely on indigenous specialists for data outside their expanse of expertise.

- Practice make sure that you include a fairly wide range of people in the initial sample. Include men, women, rich, poor, loftier- and depression-caste.

Modifications

By changing the diction of the question, you can use a similar approach to identify other types of people or relationships. For instance:

- "If yous need some communication, who do yous go to?"—This helps to identify opinion leaders.

- "Who do you most often talk to in the village?" This helps to identify social networks.

- "Is there anyone in the village who you disagree with on (topic X) ?"

—This helps to identify a range of opinions.

Notation:

AIthough certain people may have a reputation for their skills they are not necessarily the all-time informants. The success on which their reputation is built might reflect their reduced need to brand compromises rather than their skills—often wealthier people who have more land and access to college inputs and therefore are less dependent on indigenous knowledge (adapted from Fairhead in HED 1991).

Diagrams

Source: http://www.nzdl.org/gsdlmod?e=d-00000-00---off-0hdl--00-0----0-10-0---0---0direct-10---4-------0-0l--11-en-50---20-about---00-0-1-00-0--4----0-0-11-10-0utfZz-8-10&cl=CL2.8.2&d=HASH26eb53763b3685458b1c07.2.2.1>=1

Posted by: fungunpleted.blogspot.com

Share this post