A research problem is a perplexing or troubling condition. The purpose of the research is to "solve"
the problem - or to contribute to its solution - by accumulating relevant information.
Example: "The side effects of chemotherapy"
A problem statement articulates the problem to be addressed. It is something that requires
Variables and relationships are terms used by quantitative researchers.
Example: if you want to know if a new product speeds wound healing, you will conduct a
quantitative research study.
The type of problem selected for study gives direction for the method (qualitative or quantitative).
Example: Children who eat breakfast will obtain higher grades in mathematics than those who
Research hypothesis: a tentative prediction about the relationship between two or more variables
in the population under study. They are statements of actual expected relationships between
Findings from quantitative studies lead to a hypothesis.
Used in some types of quantitative research.
Should be based on sound rationale.
How do Hypotheses Help?
Foster critical thinking
Facilitates interpretation of data
Minimizes acceptance of "accidental" findings
States expected relationship between independent variable and dependent variable
Types of Hypotheses
Example: Self-reported job satisfaction predicts intentions expressed about working as a U.K.
Simple: predicts relationship between one independent variable and one dependent variable.
Example: Self-reported job satisfaction at earlier time-points (6 months, 18 months) predicts
working as a U.K. nurse at 18 months and 3 years after qualifications.
Complex: predicts relationship between two or more independent and/or two dependent.
Example: The incidence and degree of severity of subject discomfort will be less after a
30 second injection of Heparin compared to a 15 second injection.
A directional hypothesis is one that specifies not only the existence but also the expected
direction of the relationship between variables.
Example: There will be a difference in fatigue between two groups of caregivers of
preterm infants at one month post-discharge. (Note: researchers did not say if
caregivers would be more or less tired, just that there would be a difference).
A non-directional hypothesis does not stipulate the direction of the relationship.
For statistical analysis, the logic of statistical inference requires that hypotheses be
expressed as though no relationship were expected. The basis of statistical analysis.
Researchers analyze data from this philosophical perspective then see what the data
Example: mother's exercise levels postpartum are unrelated to their weight retention.
The variables are assumed to be "innocent" of any relationship until they can be shown
to be "guilty" through statistical procedures.
Null hypotheses state that there is no relationship between the independent variables and the
Hypotheses can be directional or non-directional.
Hypotheses are never proved (or disproved) through hypothesis testing; rather, they are accepted or
Winter Week 1 - Review Quantitative Content/Levels of
Research Page 1
Findings are always tentative . Example: “Findings from this study suggest….
Quantitaive Design Review
Manipulation: Experimenters do something to participants in the study.
Control: Experimenters introduce controls, including the use of a control group.
Randomization: Experimenters assign participants to control or experimental groups
To qualify as an experiment, a research design must have three properties:
Example: investigating the effect of physical exertion on mood in healthy adults.
Pretest-posttest (before-after design) involves the observation of the dependent variable at two
points in time: before and after the treatment.
Example: testing the impact of completing a questionnaire about blood donation on
subsequent blood donation behaviour in experienced blood donors: the two dependent
variables measured are the number of registrations at each blood drive and number of
successful blood donations. Measurements are made 6 months and 1 year after the
questionnaire is completed.
Posttest-only (or after only) design involves randomizing subjects to different groups and
subsequently measuring the dependent variable.
Repeated-measures design is a design in which one group of subjects is exposed to more than one
condition or treatment in a random order; also called a crossover design.
Clinical trials re studies designed to assess the safety and effectiveness of a new clinical treatment,
sometimes involving several phases, one of which (Phase III) is a randomized clinical trial using an
experimental design and, often, a large and heterogeneous sample of subjects.
An example of the experimental method:
An intervention study in which subjects are not randomly assigned to treatment conditions, but the
researcher exercises certain controls to enhance the study’s internal validity.
Look a lot like experiments because they also involve the manipulation of an independent variable
(e.g. The institution of a treatment).
However, they lack either the randomization or control-group features of true experiments.
Almost identical to the before-after experimental design discussed in the previous section,
except subjects were not randomly assigned to the groups.
Example: studying the effect of primary nursing on nursing staff morale in an urban hospital.
Since the new system of nursing care is implemented on nursing staff throughout the hospital,
randomization of nurses is not possible. Therefore, comparison data is collected from nurses in
a similar hospital that is NOT instituting primary nursing. Data is gathered on staff morale from
both hospitals before implementing the primary system (the pretest) and after its
implementation (the posttest).
The nonequivalent control-group before-after design involves two or more groups of subjects
observed before and after the implementation of any intervention.
Research Page 2
Winter week 1 - review quantitative content/levels of. A research problem is a perplexing or troubling condition. The purpose of the research is to solve the problem - or to contribute to its solution - by accumulating relevant information. A problem statement articulates the problem to be addressed. Variables and relationships are terms used by quantitative researchers. The type of problem selected for study gives direction for the method (qualitative or quantitative). Example: if you want to know if a new product speeds wound healing, you will conduct a quantitative research study. Research hypothesis: a tentative prediction about the relationship between two or more variables in the population under study. They are statements of actual expected relationships between variables. Example: children who eat breakfast will obtain higher grades in mathematics than those who don"t. Findings from quantitative studies lead to a hypothesis. States expected relationship between independent variable and dependent variable.