A. Probability Sampling Methods
A. Probability Sampling Methods
A. Probability Sampling Methods
Simple Random Sampling – ensures that each element of the population has an equal and
independent chance of being chosen.
o Identify the sample population and list all the elements of the population (sampling
frame)
Stratified Random Sampling – population is divided into subgroups or strata, according to some
variable/s of importance.
After this, a simple random sample is taken from each of the subgroups.
o Proportional stratified
o Disproportional stratified
Cluster Random Sampling – large groups or samples become the sampling units.
Systematic Random Sampling – sample is taken from every kth element of the population.
– sample elements are chosen from the population by non-random methods. More likely to produce
biased samples.
Quota sampling – similar to stratified random but selection not random. Basis of stratification is
determined by the researcher.
1. Longitudinal study – follows the subject over a period of time (6 months or more). More
accurate study of changes that occur over time.
a. cohort study – persons are studied who have been born during a particular time
period.
2. Cross-sectional study – examines the subjects at one point in time. Less expensive and easier
to conduct.
Pilot study: Miniature, trial version of the planned study. Can prevent a researcher from
conducting a large-scale study that might be an expensive disaster.
Objectives – to examine issues related to the design, sample size, data collection procedures
and data analysis approaches.
*The choice of data collection method is determined by the study hypothesis/research questions of the
study.
Criteria for selection of data Data collection methods:
collection instrument 1. Questionnaire
i. Practicality of the 2. Interview
instrument
3. Observation
ii. Reliability of the 4. Physiological measurement
instrument – consistency
5. Attitude scales
and stability
6. Psychological tests
iii. Validity of the instrument
7. Delphi technique
– ability mto gather data
that it is intended to 8. Visual analog scale
iv.
DATA COLLECTION METHODS Concerns the content of
1. Questionnaires – paper and pencil, self-report instrument. Guidelines in wording of questions:
d. Neutral wording
Types of questions:
Must be collectively exhaustive (all possible answers provided) and mutually exclusive
(no overlap between categories)
d. Contingency questions – items that are relevant for some respondents and not for
others. Example: If yes…
e. Filler Questions – items in which the researcher has no direct interest but are included in a
questionnaire to reduce the emphasis on the specific purpose of other questions.
Cover Letter – accompanies the questionnaire disclosing the writer’s purpose, who is sanctioning the
study, deadline of returning the questionnaire, offer to inform respondents of study result.
a. Unstructured interview – interviewer given a great deal of freedom to direct the course of the
interview.
*Probes – additional promptimg questions that encourage the respondent to elaborate on the
topic
b. Structured interviews – asking the same questions in the same order and in the same manner
of all respondents in the study. Even subtle changes in the wording of the interview may not be
permitted.
a. Structured – carried out when the researcher has prior knowledge about the phenomenon of
interest.
- Uses a checklist.
b. Unstructured – researcher attempts to describe events or behaviors as they occur, with no
preconceived idea of what will be seen.
C. Participant observer-overt – involved with the subjects openly and subjects know that
they are being observed by the same. Eg. Immersion with families while observing their day- to-
day lifestyle
D. Participant observer-covert – “plant”, “Spy”. Observer interacts with the subjects and
observes their behavior without their knowledge. Rarely ethical.
4. Physiological Measures – involve the collection of physical data from the subjects. Generally more
objective and accurate than many of the other data collection methods.
5. Attitude Scales – self-report, data-collection instruments that ask respondents to report their
attitudes or feelings on a continuum
A. Likert Scale – uses five or seven responses for each item ranging from Strongly Agree (5) to
Strongly Disagree (1).
B. Semantic Differential Scales – asks subjects to indicate their position or attitude about some
concept along a continuum between two adjectives.
6. Psychological Tests
7. Delphi Technique – uses several rounds of questions to seek a consensus on a particular topic from a
group of experts. To obtain group consensus without a face-to-face meeting.
9. Preexisting Data – use of existing information that has not been collected for research purposes.
Eg: patient’s charts.
- Plans for organizing the data should be made prior to data collection. Plans for analyzing
the data should also be made prior to data collection.
- Determine if questionnaires have been completed correctly. What to do with missing data.
Audio tapes transcribed.
1. Nominal – objects or events are named or categorized (Religion, gender, marital status)
1.Frequency distribution – simply counting the occurrence of values or scores represented in the data.
If range of score is less than 20, each score can be listed individually, when the range is large you can
group them into “class intervals”.
2. Graphic Presentations – have visual appeal that may cause readers to analyze the data more closely
MEASURES OF CENTRAL TENDENCY
– statistics that describe the average, typical, or most common value for a group of data.
1. Mode – category or value that occurs most often in a set of data under consideration.
3. Mean – the average sum of a set of values found by adding all values and dividing by the total
number of values.
MEASURES OF VARIABILITY
1. Range – distance between the highest and lowest value in a group of values or scores.
2. Percentile – a datum point below which lies a certain percentage of the values in a frequency
distribution.
3. Standard Deviation – Indicates the average deviation or variation of all values in a set of values from
the mean value of those data.
MEASURES OF RELATIONSHIPS
1. Correlation coefficients – pairing the value of each subject on one variable with the value on another
variable
2. Scatter plots (scatter diagram, scattergram) - graphic representation of the relationship between two
variables
(X and Y axis)
- Made in light of the study hypothesis or research question and the theoretical framework
- The final step in the research process and yet the most important one for nursing. No matter
how significant the findings may be, they are of little value to the nursing profession if not
communicated to other colleagues.