Scribd
Upload
23 views29 pages

Biostatistics Unit 6. Sampling

Sampling in statistics

Uploaded by

linetmuthoniw4
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
23 views29 pages

Biostatistics Unit 6. Sampling

Sampling in statistics

Uploaded by

linetmuthoniw4
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 29

Sampling

.
Definition

Process of choosing a representative portion of the entire population


Terminologies

1. Sample. Subset of the population that is selected for the study. Also called
subjects or respondents of the study
2. Element. Most basic unit about which information is collected
3. Representativeness. Means that the sample must be like the population in as
many ways as possible. The accessible population must be representative of
the target population
Sample and population
The population

Population is an accessible group of people who meets a well defined set of


eligibility criteria.
Consists of the totality or aggregate of the observation with which the researcher is
concerned.
Types of population.

1. Target population – is a group of individuals who meet the criteria


2. Subject or respondent population- refers to a group of individuals participating
in the study.
3. Strata or stratum – is described as a mutually exclusive segment of a population
established by one or more characteristics.
Cont

Sampling unit. Refers to specific place or location which can be used during
sampling process Sampling frame. Describes the complete list of sampling units
from which the sample is drawn.
Sampling criteria: Refers to the essential characteristics of a subject or respondent
such as ability to read and write responses on the data collecting instruments
Eligibility criteria: A description chosen by the researcher to define which elements
should be included in or excluded from the population. Such criteria may include
sex, age, marital status, education level, and diagnosis.
The steps involved in sampling.

1. Identify the target population.


2. Identify the subject or respondent population.
3. Specify the criteria for subject or respondent selection.
4. Specify the sampling design.
5. Recruit the subjects.
Sample size

The researcher must first determine the size of the sample


It can be determined using Slovin’s (1960 ) formula.
n = N/1 + Ne2
Where:
n is sample size
N is population size
e is margin error (0.05)
1 is a constant value
Cont

Use Slovin's formula to find out what sample of a population of 1,000 people you
need to take for a study.
Sampling techniques

Major categories of sampling


1. Probability (random) sampling. Each subject has a known probability of being
selected.
2. Non probability (non-random). sampling. Probability of selecting any particular
member is unknown
Probability sampling

1. Simple random sampling.


2. Systematic sampling.
3. Stratified sampling.
4. Cluster sampling.
5. Multistage area sampling.
Non probability sampling

1. Convenience sampling. Sample is selected from elements of a population that are


easily accessible
2. Purposive/Judgment sampling. You chose who you think should be in the study
3. Quota sampling. You select a predetermined number or proportion of units, in a non-
random manner
4. Snowball sampling (chain sampling or network sampling). Existing subjects provide
referrals to recruit samples required. begins with one or more study participants. It
then continues on the basis of referrals from those participants. This process
continues until you reach the desired sample, or a saturation point.
What is the appropriate sample design?

1. Degree of accuracy
2. Resources
3. Time
4. Advanced knowledge of the population
5. National versus local
6. Need for statistical analysis
Probability sampling

Simple random sampling.


Each member of the population has equal chance of being included in the samples.
Most commonly used method is the lottery or fish bowl technique.
In using lottery method there is need for complete listing of members of the
population.
The names or codes are written in pieces of paper cards and placed in a container.
The researcher draws the desired number of sample from the container.
The process is easy for small population but difficult and time consuming for large
population.
Sampling error formula

Sampling error = Z × n ​ σ / square root of n

Where:
Z = Z score value based on the confidence interval (approx. = 1.96)
σ = Population standard deviation.
n = Size of the sample​.
Conclusion

1. Probability samples are the best


2. Ensure. Representativeness, Precision
Errors in sample. Sampling error

Occurs when the sample used in the study does not represent the entire
population.
1. Systematic error (or bias). a consistent or proportional difference between the
observed and true values. (eg. a miscalibrated scale consistently records
weights as higher than they actually are)
2. Population-specific error. when a researcher doesn't understand who to
survey.
3. Selection error. when the survey is self-selected, or when only those
participants who are interested in the survey respond to the questions
4. Sampling error (random error). Sample frame error. when a sample is selected
from the wrong population data.
5. Non-response error. Inaccurate response, selection bias.
Primary and secondary data

.
Terminology definitions

Primary data. Data collected by the original user who also actively take part in
collection of the data
Secondary data. Data collected from records. This is also data collected by persons
not the users of the data at the time it collected from the records and used for
different purpose.
Numerical data. Date collected in form of numbers or figures.
Categorical data. Data collected in form of categories or description of the data in
question.
Grouped data. Data that has been grouped in frequencies.
Ungrouped data. Data that is taken the way it is collected without grouping.
Primary data

Data collected by the original user who also actively takes part in collection of the
data. Primary data is first-hand information, original in nature collected from a unit
or individual respondent directly for a predetermined purpose with active
involvement of the researcher.
Methods of collection primary data

1. Observation.
2. Interview
3. Questionnairre
4. Sampling.
5. Focused group discussion (FGD).
6. Personal enquiry.
7. Snowballing.
Secondary data

Data collected from records. This is also data collected by persons not the users of
the data at the time it collected from the records and used for different purpose
These are second hand information, statistically processed, obtained from an
already collected.
Source initially collected for some purpose and are now available for the present
purpose.
Sources of collecting secondary data

1. Official publication by government or provincial government.


2. Semi officials by municipalities, central bank, district board.
3. Publications by research institution or universities like KEMRI, AMREF, KARI
4. Reports i.e. monthly, annual, Red Cross, business commerce, etc.
5. Different journal and periodicals.
6. Report from health records and information office
Primary data vs secondary data

Primary data Secondary data


1. It is time consuming. 1. It is time saving which could have
2. Low quality and voluminous in been used on collecting data.
nature. 2. Provide larger and quality data.
3. Can be more expensive as 3. May be less expensive as compared
compared with secondary data. to primary data.
4. The researcher is prone to risk 4. The researcher may not be involved
while collecting the data. in risk as in the primary data
5. Designed for understanding and 5. It is possible to conduct a new
solving the research problem at survey that can adequately capture
hand past change and or development.
Cont

Primary data Secondary data

6. Collection process is participatory. 6. Individualized data collection Can be


Researcher is involved out dated
Numerical data Vs Categorical data

Numerical data Categorical data


1. Are data that exist in numerical form, 1. Are data that take a finite set of
such as height, number of children in a values and that can be either numerical
household and annual income. or categorical. e.g. Eye color.
2. Mostly quantitative (Number of 2. Categorical data deals with data that
patients in the ward, heights of person). can be divided into groups. Eg. Gender
3. Numerical data can be either group.
continuous i.e. number of rooms, 3. Categorical data is not countable
persons , a person’s shoe size, No. of unless they are grouped
children in a house.
Cont

Numerical data Categorical data

4. Mostly quantitative. Eg. Number of 4. Mostly qualitative. Eg. All admitted


admitted patient. patients.
5. The number of data is countable.
Numerical data Can sometimes be 5. The number of data is countable.
converted to categorical data Categorical data can sometimes be
converted to numerical data
The end

Thank you

You might also like