Chapter 2: Scientific Method Flashcards
What characteristics should research procedures be based on?
-self-correcting research characteristics
-findings must not be derived from biased research
-findings must be made public
-experiment’s must be reproducible by other scientists at later time
-experiments must be empirical
-science should be predictive
Self correcting research characteristics
if the results of a research study are later found to be false, the research should be conducted again so that the conclusions or results may be modified. When conducting research the truth is not found in one experiment or study but often requires many studies EX: Acupuncture
findings must not be derived from a biased research
requires objectivity
should not be any bias based on the researchers personal beliefs, perceptions, values or emotions
researcher must develop rules and procedures for the research (specific hypothesis, research question, significance level)
True or False: objectivity in research is met fairly easy in quantitative research
true
findings must be made public
most researchers attempt to publish their findings in peer-reviewed professional journals, may be made public by mass media but peer review is essential
reproducible by other scientists at later time
requires other scientists to replicate the same research process using the same research design and methodology
True or false: if research is experimental, the researcher may manipulate a variable and then observe the results
true
science should be predictive
such predictions are demonstrated in scientific theories that arise as a result of research studies. The predictions allow for furthur research that tests the theories
Types of research design
cross-sectional
retrospective
prospective
longitudinal
Cross-sectional
a study of a group of people at a given point in time. It could be carried out by administering a survey on a particular date in a given year
Retrospective
a study involving past records of a group or groups over a long period of time (years) to assess risk factors of a disease such as lung cancer
Prospective
could involve studying a group over a long period of time (years) to assess risk for getting a disease sometime in the future such as diseases as a result of smoking cigarettes
Longitudinal
This is a study of the same individuals (cohort group) over a long period of time on the same health variables and risk factors
Descriptive design
may include surveys (quantitative) or qualitative studies
Analytical studies
quantitative and are classified as observational, experimental and quasi-experimental type studies
Predictive statistics
also called inference studies
assess the cause and effect of variables
Methods used to conduct research studies
case studies-observation of people plus interviews
laboratory experiments-conducted in a controlled environment
epidemiological studies: Analyze data from various population groups over a point in time or for years
controlled clinical trials: may involve a number of people using an experimental group or a control group
Surveys; Pen and pencil, computer, telephone, face-to-face, internet
Data collection: collecting data from archived record
Evidence-based research
-especially important for clinical practices whether it is medicine, nursing, psychology, speech-language-and-hearing or the many alternative practices offered today
-study evidence or result is integrated with clinical expertise and patient values when making decisions about patient care
Four main steps (evidence based research)
- framing the clinical question
- finding evidence
3.Assesing the evidence
4.making the clinical decision
Steps in scientific research
- Identify a problem (needs assessment)
- Review existing research based on the problem (literature review)
3.Develop either hypotheses or research questions - Plan the research design and methodology (complete with the type of statistics to be used and the statistical level of evaluation)
- Collect and analyze the data
- Describe the results of each hypothesis or research question
- Publicize the results in review journals
Problems with health information
-sources not reliable
-nonprofessionals or pretend scientists promoting products
-some medical professionals promoting products for their pwn financed gain
media hype on certain products and drugs or supplements
Validity
the extent to which the test predicts the outcome it is supposed to predict
reliability
test has it if it yields consistent results
Ways to investigate validity
- check for verification of the product or drug, such as a peer-review article or report
2.Investigate safety research on products and side effects of medicines/drugs - Don’t rely on the results of one study, remember that several studies are often required
- If you read a report, investigate the origin of the report. Find out of it was from a peer-reviewed study or medical institution. If not, be wary.
- Ask your doctor
- Read reliable magazines and newsletters
- Obtain info from reputable sources
