Research – Discussion 6 – Descriptive Research One of the skills a good researcher must have is the ability to interpret data (whether qualitative or quant
Research – Discussion 6 – Descriptive Research One of the skills a good researcher must have is the ability to interpret data (whether qualitative or quantitative) accurately. As a good researcher, what is the difference between correlation and causation? Provide some examples in your answer QUANTITATIVE RESEARCH
METHODOLOGIES
Descriptive Research Chapter Eight
What is Descriptive Quantitative
Research?
This general category of research designs involves
either identifying the characteristics of an observed
phenomenon or exploring the possible associations
among two or more phenomena.
In every case, descriptive research examines a
situation as it is.
Descriptive Research Designs
All of these
studies yield
quantitative
information
that can be
summarized
through
statistical
analyses.
Observation Studies
Correlational Research
Developmental Designs
Survey Research
Observation Studies
In quantitative research, observation studies are
different than observations in qualitative research
An observation study does not necessarily involve
members of the animal kingdom
A quantified observation study tends to have a
particular, pre specified focus
When humans are the object, the behavior is quantified
in some way – behavior may be counted to determine
its overall frequency
In some situations, behavior is rated
The researcher strives to be as objective as possible
Observation Studies
Strategies to maintain objectivity:
Define the behavior precisely and concretely
Divide the observation period into small segments
Use a rating scale to evaluate the behavior
Have two or three people rate the same behavior
independently
Train the rater(s) to use specific criteria when counting
or evaluating
Correlational Research
A correlational study examines the extent to which
differences in one characteristic or variable are
related to differences in one or more other
variables
A correlation exists if, when one variable increases,
another variable increases or decreases in a
somewhat predictable fashion
Researchers gather data about two or more
characteristics for a particular group of people or
other appropriate units of study
Correlational Research
Whatever the nature of the data, at least two different
characteristics are measured in order to determine
whether and in what way these characteristics are
interrelated
A scatter plot may be used to allow a visual inspection
of the relationship between two variables
The scatter plot allows us to describe the homogeneity or
heterogeneity of variables
It also allows us to describe the degree to which the two
variables are intercorrelated, perhaps by computing a
statistic known as a correlation coefficient
It allows us to interpret data and give them meaning
Correlational Data – Caution
In all correlational studies, be alert for faulty logic
We can never infer a cause-and-effect relationship on
the basis of correlation alone
Correlation does not, in and of itself, indicate
causation
A way to explore the dynamics beneath the correlation
is through a statistical procedure called structural
equation modeling (SEM)
Another way is to follow up a correlational study with
one or more experimental studies to test various
hypotheses about what causes what
Developmental Designs
There are two types of Developmental Designs
A cross-sectional study, wherein people from several
different age-groups are sampled and compared
A longitudinal study, wherein a single group of people
is followed over the course of several months or years,
and data related to the characteristic(s) under
investigation are collected at various times
Developmental Designs
When longitudinal studies are also correlational studies,
they enable researchers to identify potential mediating
and moderating variables in correlational relationships
A mediating variable (intervening variable) may help
explain why a characteristic is correlated with another
characteristic
A moderating variable influences the nature and strength of
a correlational relationship
A statistical analysis such as structural equation modeling
(SEM) can be helpful for identifying mediating and
moderating variables in a longitudinal study
Correlational studies cannot conclusively demonstrate
cause-and-effect relationships
Developmental Designs
Cross-sectional studies are easier and more
expedient to conduct than longitudinal studies
because the researcher can collect all the needed
data at a single time
Disadvantages of cross-sectional studies
Different age groups sampled may have been raised
under different environmental conditions
We cannot compute correlations between
characteristics at different age levels
Developmental Designs
To address some of the weaknesses of longitudinal
and cross-sectional designs, researchers occasionally
combine both approaches in what is known as a
cohort-sequential study
Cross-sectional, longitudinal, and cohort-sequential
designs are used in a variety of disciplines, but are
most commonly seen in developmental research
studies
Survey Research
Survey research involves acquiring information
about one or more groups of people by asking
them questions and tabulating their answers
The ultimate goal is to learn about a large
population by surveying a sample of that
population; thus, this approach might be called a
descriptive survey or normative survey
Survey Research
A survey is simple in design; however, survey design
makes critical demands on the researcher
Survey research
Captures a moment in time
Relies on self-reporting of data (by individuals)
May employ
Face-to-face interviews
Telephone interviews
Written questionnaire
Planning for Data Collection in a
Descriptive Study
In quantitative research, a descriptive study
invariable involves measuring one or more variables
in some way
Consider the distinction between substantial and
insubstantial phenomena
Many descriptive studies address complex variables
using a variety of strategies
Checklist
Rating scale/Likert scale
Choosing a Sample in a Descriptive
Study
A researcher will usually not study an entire
population – instead, he or she will select a subset,
or sample, of the population
The researcher can use the results obtained from the
sample to make generalizations about the entire
population only if the sample is truly representative
of the population
Sampling Designs
Two
categories of
sampling:
Probability
Sampling and
Non-
Probability
Sampling
Probability sampling
Simple random sampling
Stratified random sampling
Proportional stratified sampling
Cluster sampling
Systematic sampling
Sampling Designs
Two
categories of
sampling:
Probability
Sampling and
Non
Probability
Sampling
Nonprobability Sampling
Quota sampling
Purposive sampling
Sampling Surveys of Large Populations
The sample must be representative of the
population about which inferences are to be made
Random does not mean haphazard
Identifying a Sufficient Sample Size
The larger the sample size, the better
The sample size is referred to by the symbol N
Guidelines
For smaller populations, say, N=100 or fewer, survey
the entire population
If N is around 500, 50% should be sampled
If N is around 1500, 20% should be sampled
If N=5000, a sample size of 100 should be adequate
Bias in Research Sampling
Bias is any influence, condition, or set of conditions
that singly or in combination distort the data
Bias attacks the integrity of the facts
Report response rate
Acknowledging the Probable Presence
of Bias
When formulating conclusions about the data, a
researcher must be sure to consider the effect that
bias may have had in distorting the data
Interpreting Data in Descriptive
Research
There is more to research than data collection
All research activity is subordinate to the research
problem itself
Two basic principles of research
Seek the answer to a problem in light of the data that
relate to the problem
Interpretation of the data is all-important
