Research Design and Methods
Dr. Emmanuel Kwesi Arthur
Department of Materials Engineering,
College of Engineering,
Kwame Nkrumah University of Science and Technology
© February 2019
Lecture Six
What is Experimental design - the Scientific Method?
Experimental Design is a logical, consistent process for stating and solving problems in the natural world.
What are the steps to follow?
Generally-1. Observe 2. Formulate a Question…Problem
Statement3. Research or Infer to formulate a
Hypothesis4. Design a Procedure to test this
hypothesis5. Experiment and record data6. Analyze the Results7. Draw a Conclusion and
communicate the results
The Observation starts it all…
An observation is a visible or provable fact or occurrence
VS.An inference is, “the act of reasoning from factual knowledge or evidence.”This is your opinion drawn on the observations you have made.
Observation vs. Inference
Observation
Observation
Inference
Careful observations lead to questions that arise…
A Problem Statement is a question that compares two variables.
Example: Does the change in the length of daylight affect the leaf color of deciduous trees?
Collecting DATA
QUALITATIVE
This is a WORD or “quality” – a subjective measure other than an number…Examples: An odor, color,
texture, taste, etc.
QUANTITATIVE
This is a NUMBER or “quantity” – an objective measure or observation…Examples: Distance, mass,
volume, density
http://images.google.com/imgres?imgurl=http://www.amcgltd.com/archives/dude-smell-this.jpg&imgrefurl=http://www.amcgltd.com/archives/cat_cats.html&usg=__5jAJ8tTJSZ2XToRS_0p94ckkbH8=&h=424&w=540&sz=52&hl=en&start=2&um=1&tbnid=8fHyrQRLWglcHM:&tbnh=104&tbnw=132&prev=/images?q=smell&hl=en&rls=com.microsoft:en-us&sa=N&um=1http://images.google.com/imgres?imgurl=http://www.amcgltd.com/archives/dude-smell-this.jpg&imgrefurl=http://www.amcgltd.com/archives/cat_cats.html&usg=__5jAJ8tTJSZ2XToRS_0p94ckkbH8=&h=424&w=540&sz=52&hl=en&start=2&um=1&tbnid=8fHyrQRLWglcHM:&tbnh=104&tbnw=132&prev=/images?q=smell&hl=en&rls=com.microsoft:en-us&sa=N&um=1
Research Sample
Population entire group about whom conclusion drawn
Sample portion of population actually observed
Representative Sample characteristics similar to population
opposite of “biased sample”
Random Sample equal chance of being selected
Sampling
Target Pop.
(N)
Sample (n)
Effective Sampling produces a n which is representative of N
Note: n is only ever representative of the N it was drawn from, i.e. not necessarily the general population.
Sampling
Statistics
The dependent variable can be generalised from n to N
In sampling, we gather data on an entire “population”by measuring only a subset of that population, knownas the sample.
A population consists of all of the individual elementsin a defined area.
Sampling Design
Are there too many people in the group that you arestudying?
Are you limited in time and resources?
If you answered yes to one or both questions, youmight want to select a sampling design to carry outyour study.
Sampling Design
A simple random sample is a selection of individuals chosen so that each point in the population has an equal chance of being selected.
Each item in a “population” can be assigned a number. Then the simple random sample can be selected by using a random number table or a random number generator (using a computer).
Sampling Design
A well-defined sample has the same characteristics as the population as a whole
It is very important to:
define the population before selecting the sample
decide the size of the sample.
How big should a sample be?
The bigger the sample size the greater will be its accuracy.
Once a researcher decides on a sample, he needs to obtain data from this sample.
Sampling Design
The data were collected using an internet questionnaire survey. Six hundred Saudi engineering companies were selected from 2,002 companies obtained from the Chamber of commerce database.
Determine the used research method, sample and population in the above statment?
Example
What quant researchers worry about
Is my sample size big enough?
Have I used the correct statistical test?
Are my results generalisable?
Are my results/methods/results reproducible?
Am I measuring things the right way?
What’s wrong with quant research?
Some things can’t be measured – or measured accurately
Doesn’t tell you why
Can be impersonal – no engagement with human behaviours or individuals
Data can be static – snapshots of a point in time
Can tell a version of the truth (or a lie?) “Lies, damned lies and statistics” – persuasive power of numbers