Research Exam 2

What is a research problem?

A obscure or troubling condition. This is basically to solve the problem or contribute a solution by gathering relevant data.

What is a problem statement?

This will express the problem and offer a argument by explaining the need for such study.

What is non-experimental research?

This is observational and a descriptive study

None intervenes or manipulates the independent variable

We want to know the r/s between the variables

IV are gender, smoking these cannot be manipulated!

What is descriptive correlational?

This is to describe variables if they are related without attributing a cause and effect connection

What is a quasi-experimental study?

This is a controlled trial but no randomization

What is experimental research?

This is a randomized controlled trial, the golden standard

What is a correlational design?

• Is associated between variables

• This is weak compared to randomized controlled trials since it is not getting manipulated

• This is a non-experimental research design

What is prospective correlational design?

This is too a correlational design meaning to see a potential cause in the present, but this is too a non-experimental design. Done in a cohort study if someone who has a miscarriage or not would later get depression in 6 months.

Stronger than retrospective designs

What is retrospective correlational design?

an outcome in the present (e.g., depression) is linked to a hypothesized cause occurring in the past (e.g., having had a miscarriage).

This is done with a case-control design meaning that these cases (e.g., those with lung cancer) are compared to “controls” (e.g., those without lung cancer) on prior potential causes (e.g., smoking habits)

What is a cohort study?

This is defined by a group of people is followed over time to check associations between different interventions and its outcomes.

What is within-subjects design?

When the same people are compared at different times or under different

conditions (quasi-experimental)

What is between subjects’ design?

When different people

are compared (e.g., men and women).

What is a prospective cohort?

This is recruiting subjects b4 any intervention and follows them into the future.

Cohort

What is a retrospective cohort?

Will identify subjects from past records what they got and follow them till now.

• Case control

What is a statement purpose?

This will summarize an overall goal.

What is a hypothesis?

This is a prediction which involves an expected r/s between 2 or more variables. Independent Variable and Dependent Variable

What is the independent variable?

The presumed cause or influence within a population

What is the dependent variable?

The presumed outcome or effect within a population

In qualitative research, do they use hypotheses?

Not really because they want the investigation to be guided by the participants viewpoints.

What makes up a good research problem for quantitative research?

It involves concepts that are well-developed and involves in measurement.

What makes up a good research problem for qualitative research?

That it develops a rich dependent situation to better understand a poorly understood phenomenon.

Is this research question qualitative or quantitative? “Exploring the meaning or experience of depression among people with chronic illness?”

This is qualitative because you are exploring meanings or personal experience.

Is this research question qualitative or quantitative? “Whether people with chronic illness are more likely to develop depression than people without a chronic illness?”

This is quantitative because you are able to measure levels of depression in people with or without a chronic illness.

What is a directional hypothesis?

Specifies the expected direction of the relationship between variables (quantitative)

• Potentially biased b/c research is already emphasizing an anticipated outcome

• Deductive is NOT this! It is Inductive

What is a nondirectional hypothesis?

Does not stipulate the direction of the relationship (qualitative)

Deductive is NOT this! Inductive is!!

What is an alternative hypothesis?

There is an effect or difference in the population in general

What are null hypotheses?

Does not stipulate the direction of the relationship (this is rejected)

In example: “Older patients and younger patients are equally likely to fall.”

In statistical testing, underlying null hypotheses are assumed, without being stated.

Is this a null or normal hypothesis? “Relaxation training would have a significant effect on the physical and psychological health status of post myocardial infraction (MI) patients”?

This is a normal if not directional hypothesis because it explains there is a significant difference in the effect on the health status on MI patients.

Is this a null or a normal hypothesis? “Oxygen inhalation by nasal cannula of up to 6L/min does not affect oral temperature measurement taken with an electronic thermometer”?

This is a null hypothesis because it says DOES NOT AFFECT the oral temperature!

A cross-sectional design, when is data collected?

Data are collected at a single point in time

A longitudinal design, when is data collected?

Data are collected two or more times over an extended period.

What is attrition?

Loss of participants over time

An issue in longitudinal studies

What is a quasi-experimental study?

Involve an intervention but lack either randomization or a control group. The lack of

randomization cannot assure that the groups are equal.

This uses within subjects design.

What are the two categories that fall in with quasi-experiments?

• Nonequivalent control group designs

• Within-subjects designs

What does nonequivalent control group designs mean?

Those getting the intervention are compared with a nonrandomized comparison group.

This is weaker

What does pre-test mean?

can provide some information as to the equivalence of the group at the beginning of the experiment. (making certain people worried due to the changes AND is more expensive)

What does a one-group pre-test and post-test design mean?

yield extremely weak evidence of causal relationships.

What is a pre-and post-test design?

Involves observing the outcome before and after the intervention (design).

Collects baseline (pretest) data on the DV before intervention and post-test (outcome) data after it.

Quasi and quantitative experiments

What is a post-test design only?

Involves randomizing people to different groups and then measuring outcomes after the intervention

What is a true RCT?

Are the “gold standard” for therapy questions because they yield the most persuasive evidence about the effects of an intervention. Through randomization to groups, researchers come as close as possible to attaining an ideal counterfactual.

What is a cofounding variable?

are extraneous variables that the researcher failed to control or eliminate damaging the internal validity of an experiment that the results you get do not reflect the actual

relationship between the variables under investigation.

What are 4 types of extraneous variables?

Demand characteristics, Experimenter / Investigator Effects, Participant variables, Situational variables

What does demand characteristics extraneous variable mean?

Environmental clues which tell the participant how to behave, like features in the surrounding or researcher’s non-verbal behavior.

What does experiment/investigator variable mean?

Where the researcher unintentionally affects the outcome by giving clues to the participants about how they should behave.

What does participant variable mean?

Like prior knowledge, health status or any other individual characteristic that could affect the outcome.

What does situational variable mean?

Like noise, lighting or temperature in the environment.

What does counterfactual mean?

is what would have happened to the same people exposed to a “cause” if they

simultaneously were not exposed to the cause.

What does control mean in quantitative research?

Is what would have happened to the same people exposed to a “cause” if they

simultaneously were not exposed to the cause.

What does mask or blinding mean?

Critical information is being withheld to avert bias.

What is a cross-over design?

Subjects are exposed to 2+ conditions in random order. Subjects serve as their own control.

What does reliability?

Necessary but not sufficient for validity

Something can be reliable but not valid. Something cannot be valid without being

reliable.

What are the types of validity mean?

Statistical conclusion validity, Internal validity, External validity, Construct validity

What does Statistical conclusion validity?

the ability to detect true r/s statistically

Example: Is there a relationship between variables?

What does Internal validity?

The extent to which it can be inferred that the independent variable caused or influenced the dependent variable

Example: Is the relationship causal?

What does External validity?

External validity—the generalizability of the observed r/s across samples, settings, or time

Example: Can the findings be generalized to other groups, times,

and settings?

What does construct validity?

Construct validity—the degree to which key constructs are adequately captured in the study.

Example: What are the cause-and-effect constructs?

What is attrition bias?

Essentially is a selection bias that occurs after the study unfolds

What is selection bias?

The most challenging threat to the internal validity of studies not using an experimental design, but it can be partially addressed using control strategies described in the previous section.

What is a good sample?

Selection of a portion of the population (a sample) to represent the entire population

What is sampling bias?

overrepresenting or underrepresenting population segment in terms of

key characteristics.

What do strata mean?

subpopulations of a population (e.g., male/female)

What is a representative sample?

A sample whose key characteristics closely approximate those of the population—a

sampling goal in quantitative research.

What is probability sampling?

Involves random selection of elements: each element has an equal, independent chance of being selected.

The types are: simple random sampling, stratified random sampling and systematic sampling

What does simple random sampling mean?

Researchers establish a sampling frame—a list of

population elements.

What does stratified random sampling mean?

The population is first divided into two or more strata, from which elements are randomly selected.

What does systematic sampling mean?

Involves the selection of every kth case from a list, such as every 10th person on a patient list

What nonprobability sampling?

Does not involve selection of elements at random; is rarely representative of the

population. The types are Convenience sampling, Quota sampling, Consecutive sampling and Purposive sampling.

What is convenience sampling?

selecting the most conveniently available people as participants

What is quota sampling?

identifying population strata and figuring out how many people are needed from each stratum

What is consecutive sampling?

Recruiting all people from an accessible population over a specific time interval

What is purposive sampling?

handpicking sample members

What is a power analysis?

Researchers can estimate how large their samples should be for testing hypotheses

• Must estimate how large the group difference will be on a key outcome

• Sample size needs can and should be estimated

What if they do not use power analysis?

Estimates sample size needs. Large samples are preferable because they enhance statistical conclusion validity and tend to be more representative, but large samples do not guarantee representativeness.