research design interactive presentation. directions to move forward, click this arrow at the bottom...

39
Research Design Interactive Presentation

Upload: marilynn-gaines

Post on 25-Dec-2015

222 views

Category:

Documents


3 download

TRANSCRIPT

Page 1: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Research Design

Interactive Presentation

Page 2: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Directions

To move forward, click this arrow at the bottom right of the screen.

To go back, click this arrow at the bottom left of the screen.

To answer a question, click the button with the correct answer. A B C D

Page 3: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Two Main Types of Research Design

• Experimental Studies– Studies that involve the control and

manipulation of variables• Observational Studies (i.e., quasi-

experimental studies)– Studies that involve the observation of

individuals or the measurement of certain outcomes. No attempt is made to affect the outcome (e.g., no treatment is given).

Page 4: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Experimental and Observational Studies

Both experimental and observational studies aim to determine the cause(s) for natural phenomena. – Once the research question has been

defined, the next step is to determine which type of research design will provide you with the most accurate answer for the type of question you are asking.

Page 5: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Experimental Studies

Experimental studies are conducted in controlled conditions that allow the researcher to minimize bias and obtain valid, reliable results.– All variable measurements and manipulations

are under researcher control. For this reason, experimental studies are considered the “gold standard” of research methodology.

Page 6: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Observational Studies

Observational studies are considered less invasive than experimental studies because they do not require a change of the environment.– Observational studies are used when experimental

studies are not ethically possible (e.g., when studying the association between a risk factor and a disease).

– Observational studies include many variables that are not under researcher control and therefore are not considered as effective in answering research questions as experimental studies.

Page 7: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Difference Between Experimental and Observational Studies:

Another key difference between experimental and observational studies is that experimental studies test for causal relationships while observational studies can only test for correlation.

Causal Relationship: A relationship in which one variable can be said to be the cause of another.

Correlation: The degree of relationship between two different variables.

Page 8: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Correlation vs. Causation• Note: Correlation does not mean causation. This

means that just because two variables are related, it does not mean that one variable causes the other.– Example: There is a strong positive relationship, or

association, between men who use hair growth products and men who have hair loss. This does not mean that hair growth products cause hair loss.

• The only way to determine whether one variable causes another variable is to conduct an experimental study.

Page 9: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Purpose of Different Study Designs

• Researchers design experimental studies whenever possible. Many times ethical considerations require that an observational study be performed instead of an experimental study.

• For example, observational studies are the only option where the variable being tested may be harmful. – If a researcher wants to study the relationship between HPV and

cervical cancer, it would be unethical to administer the virus to an experimental treatment group and compare their clinical outcomes with a control group. Therefore, an observational study would be necessary to study this relationship.

– Observational studies are often used in epidemiological studies, psychological studies, and sociological studies.

Page 10: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Types of Observational Studies

• Cohort Studies (i.e., longitudinal studies)– Prospective cohort studies– Retrospective cohort studies

• Cross-Sectional Studies

Page 11: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Cohort Study

A research study that compares a particular outcome (such as lung cancer) in groups of individuals who are alike in many ways but differ by the variable the researcher is studying (e.g., female nurses who smoke compared with those who do not smoke)

Page 12: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Cohort Studies

Prospective Cohort Study: Outcomes have not yet occurred as study begins– Not always feasible as they take considerable time,

money, and resources to complete; and they require a large number of subjects

Example: Track a group of people with and without a risk factor through time to observe who develops the disease

Retrospective Cohort Study: Outcomes have already occurred as the study begins

Example: Review old medical records of a group of people with and without a risk factor and determine who developed the disease from birth to death

Page 13: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Cross-Sectional Study

• A research study that involve multiple observations (e.g., an interview or questionnaire), which may be useful to determine how variables affect each other at the same time and period.

Example: Questionnaire to determine how income is related to weight

• Cross-sectional studies can also provide a snapshot of the frequency and characteristics of a disease in a population at a particular point in time.

Page 14: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Test For Understanding

• Go to the following website and read a brief synopsis of an observational study conducted by the Women’s Health Initiative: http://www.nhlbi.nih.gov/whi/os.htm

• What type of observational study was conducted?

Prospective Cohort Study

Retrospective Cohort Study

Cross-Sectional Study

A

B

C

Page 15: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Very Good!

You are ready to move on to the next topic.

Page 16: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

SORRY! Go Back and Try Again

A retrospective study follows a cohort back in time. This study followed more than 93,000 postmenopausal women forward in time for eight years.

Page 17: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

SORRY! Go Back and Try Again

A cross-sectional study takes place at a single point in time. This study followed more than 93,000 postmenopausal women for a period of eight years.

Page 18: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Experimental Studies

In PBS you investigated the effect of the “fight or flight” response on heart rate. In HBS you investigated the effect of different factors of muscle fatigue. In MI you learned how clinical trials are designed to assess the safety and efficacy of experimental treatments. All of these have one thing in common – they are all examples of experimental studies.

Page 19: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Experimental Studies

• Research studies conducted to determine the effect that one variable has upon another variable. – Researcher maximizes control over as many aspects

of the environment as possible in order to prove causation.

– Researcher designs studies that are reliable and valid.

Reliable: A study in which another researcher can perform exactly the same experiment and generate results with the same statistical significance.

Valid: A study in which the results accurately describe the

real world.

Page 20: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Validity

Researchers design studies that minimize or eliminate systematic errors, or biases, in the data collection process. • Valid studies allow scientists to generalize

conclusions from a given study to the real world.• Biases can be minimized through the following:

– Control groups– Randomization– Blinding

Page 21: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Controlled and Randomized Studies

• Controlled Study: One group receives the treatment and another group does not.

• Randomized Study: The control group and treatment group are chosen at random.

Page 22: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Blinding

• As you learned with clinical trials, blinding (when performing studies involving humans) allows researchers to eliminate bias that participants or investigators have introduced by having knowledge of the treatment or control conditions in an experiment.

• Single-blind• Double-blind• Triple-blind

Page 23: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Single-Blind

The participant does not know whether he or she has been allocated to a treatment or control group.– Identifies the placebo effect, where

participants in the control group of an experiment exhibit some of the effects of the treatment because they think they are receiving the treatment.

Page 24: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Double-Blind

Neither the participant nor the researcher knows whether the participant has been allocated to a treatment or control group.– If a researcher knows that a participant is

going to receive a treatment, they may act more cautiously than if they were administering a control.

Page 25: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Triple-Blind

Neither the participant, the researcher, nor the response gatherer knows whether the participant has been allocated to a treatment or control group.– Prevents the response gatherer (i.e., the

person responsible for observing and measuring data in an experiment) from being influenced by membership knowledge of the treatment and control groups.

Page 26: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Confounding Variables

• In order to ensure valid results, the researcher must eliminate any confounding variables that might affect the results of the experiment.

Confounding Variable: A variable that was not accounted for in the experimental design, varies systematically with the dependent variable, and prevents a clear interpretation of the effect of the independent variable on the dependent variable.

Page 27: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Example of a Confounding Variable

A soccer coach wanted to determine whether vitamins would improve the team’s playing ability, so the coach set up an experiment. The coach assigned the offensive players into the experimental group and gave the players vitamins. He assigned the defensive players into the control group and gave them a placebo. The offensive players also ran an extra two miles a day. At the end of the two week study, the coach noted that the offensive players were playing significantly better than the defensive players.– What are the confounding variables in this

experiment?

Page 28: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

What should the coach have done differently to ensure validity?

Because of the design of the experiment, there is no way to determine whether the change in performance is due to the vitamins, running two extra miles a day, or the inherent difference between the offensive and defensive players. Both of these (running an extra two miles a day and assignment of players to an experimental group and control group) are confounding variables.– What should the coach have done differently to

ensure validity?

Page 29: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Test For Understanding

Answer Conclusion question 1 on Project 2.1.1 Scientific Research. Once you have completed this question, you may proceed to the next slide.

Page 30: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Test For Understanding

When the experimenter evaluates the outcome of a study without knowing which participants were in the experimental versus control group and the participants and data gatherers are also unaware of group assignments, then the study is said to be

Retrospective

Randomized

Triple-Blind

B

C

A

Page 31: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Very Good!

You are ready to move on to the next question.

Page 32: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

SORRY! Go Back and Try Again

A retrospective study follows a cohort back in time. This is a controlled, experimental study where all individuals involved did not know who was assigned to the experimental group and who was assigned to the control group.

Page 33: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

SORRY! Go Back and Try Again

A randomized study is an experimental study where the participants are randomly assigned to the control group or the experimental group. Although this is a controlled, experimental study, no mention of how participants were assigned to groups was mentioned.

Page 34: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Experimental vs. Observational Study?

• Why would a researcher choose to perform an experimental study instead of an observational study?

Page 35: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Experimental Study

An experimental study would allow the researcher to determine a causal relationship between the independent variable and the dependent variable. The subject the researcher is studying must allow the researcher to design a controlled environment where all confounding variables are eliminated.

Page 36: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Test For Understanding

A researcher wants to study the relationship between diet and the occurrence of breast cancer. Which type of study should the researcher conduct?

Cohort Study

Cross-Sectional Study

Experimental StudyC

B

A

Page 37: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

Very Good!

You have now completed the Research Design Interactive Presentation.

Page 38: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

SORRY! Go Back and Try Again

A cross-sectional study only involves a single observation that is useful if an immediate response to a specific question is required. Because of the nature of determining the relationship between diet and the occurrence of breast cancer, following a cohort group throughout a long period of time is the best choice to determine correlation. Unfortunately, this type of study will not allow the researcher to determine whether the relationship between diet and the occurrence of breast cancer is causal.

Page 39: Research Design Interactive Presentation. Directions To move forward, click this arrow at the bottom right of the screen. To go back, click this arrow

SORRY! Go Back and Try Again

An experimental study is only possible if the researcher can control the environment. Because of the nature of this study, there is no way that the researcher can conduct an experimental study. Following a cohort group throughout a long period of time is the best choice to determine correlation. Unfortunately, this type of study will not allow the researcher to determine whether the relationship between diet and the occurrence of breast cancer is causal.