It is based on the works of others – not a copy of their works
It can be replicated It is generalizable to other settings It is based on some logical rationale and tied to
theory It is doable! It generates new questions or is cyclical in nature It is incremental It is an apolitical activity intended for the
betterment of society
What research is and what it isn’t
NONEXPERIMENTAL
1. Descriptive – describe the characteristics of an existing phenomenon
2. Historical – relate events that have occurred in the past to current events
3. Correlational – examine the relationships between variables
TYPES OF RESEARCH
EXPERIMENTAL – Test for true cause and effect relationships
QUASI-EXPERIMENTAL – Test for causal relationships without having full control
TYPES OF RESEARCH
Ask new questions
Ask and identify
factors
Hypothesize
Gather Data
Test the Hypothesis
Reconsider the Theory
THE SCIENTIFIC METHOD
1. What areas within your discipline especially interest you?
2. Come up with 2 possible questions you could possibly work on within a period of 2 months.
3. Of the different types of research described and discussed, which one do you think best fits the type of research that is suited to answer your questions in #2?
Exercises
Significance levels Null hypotheses Independent variables Factorial designs Samples Populations
The Language of Research
VARIABLE – represents a class of outcomes that can take on more than one value example: HAIR COLOR (red, brown, blonde, …)
Variables of the same name can take on different values
example: HEIGHT (in actual measurement or in
rank)
VARIABLES
Represent the outcomes of a research study that may depend on the experimental treatment or on what the researcher changes or manipulates
example: growth rate (increment in height) of monggo seedlings with different wavelengths of light
DEPENDENT VARIABLES
Represent the treatments or conditions that the researcher controls to test their effects on some outcome
Also known as treatment variable
Must take on at least two levels or values
INDEPENDENT VARIABLES
General Rule: When the researcher is manipulating anything or assigning participants/samples to groups based on some characteristic, such as age, gender, or treatment, that variable is the independent variable.
In some cases, when the interest is not on the effects of one thing to another, but only in how variables may be related, there are no independent variables.
NOTHING IS MANIPULATED
Factorial Designs – experiments that include more than one independent variables
example: 1. Gender (male, female) 2. Age (1-3,4-6,7-9,10-12) 3. Health status (undernourished,
normal, overweight)
The best independent variable is independent of any other variable in the same study in order to contribute the maximum amount of understanding beyond what other independent variables can offer
The best dependent variable is one that is sensitive to changes in the different levels of the independent variable. Otherwise, even if the treatment had an effect, you’d never know it.
Control variables – has a potential influence on the dependent or variable and has to be removed or controlled
Extraneous variable – has an unpredictable impact on the dependent variable
Moderator variable – related to the variables of interest, masking the true relationship between the dependent and the independent
OTHER VARIABLES
“educated guesses”
Reflect the general problem statement or question that was the motivation for undertaking the research study
Provide transition from a problem statement into a form that is more amenable to testing
HYPOTHESES
a statement of equality:- There will be no difference…- There is no relationship…
Refers to the population A starting point (state of affairs accepted
as true in the absence of other information) A benchmark against which the actual
outcomes of a study will be measured or compared to see if the difference is due to chance or something else
NULL HYPOTHESIS (H0)
a definite statement of the relationship between two variables; explicit
A statement of inequality – directional (more than or less than) and nondirectional (different from)
It is this hypothesis that is tested directly (in a sample of a population) as one step in the research process
RESEARCH HYPOTHESIS (H1)
1. Stated in declarative form and not as a question.
2. Clearly describes the relationship between variables.
3. Reflect the theory or literature they are based on.
4. Should be brief and to the point.5. Testable.
Criteria for a Good Hypothesis
Traditional approach:
IDEA RESEARCH ? H1 LIT REVIEW
Better approach:
IDEA LIT REVIEW
RESEARCH ? H1
SELECTING A PROBLEM
Select a problem that you are genuinely interested in.
SELECTING A PROBLEM
1. Falling in love with your idea can be fatal.2. Sticking with the 1st idea that comes to
mind isn’t always wise.3. Doing something trivial that has no
conceptual basis or no apparent importance
4. Biting off more than you can chew.5. Doing something that has already been
done.
ATTENTION: HAZARDS!!!