1.4 Selecting a Research Method
4 min read•january 6, 2023
Sadiyya Holsey
Sadiyya Holsey
When selecting a research method, the researcher should write down their goals and what their is.
Selecting a research method depends on what a researcher wants to show or prove. For example, if a researcher only wants to show a between two variables, then an experiment would not be optimal. If a researcher wants to show between two variables, then an experiment is the best method.
When performing any research, the researcher should be sure of their results to ensure that they show true relationships. ✔️
Validity
Selecting a research method also depends on the of the experiments. There are two types of : and .
refers to how generalizable the results of the experiment are. For example, if the study on a drug is done on an Asian, middle-aged, average-weight man with high blood pressure, can the results be generalized to the population?
is when a study shows a truthful cause-and-effect relationship and the researcher is confident that the changes in the dependent variable were produced only by the independent variable. A confounding variable hurts the because it creates lower confidence in the research conclusion.
Confounding Variables
limit the confidence that researchers have in their conclusions. To recall from the last key topic, the confounding variable is an outside influence (variable) that changes the effect of a dependent and independent variable. For example, we looked at the between crime ⛓️ and the sale of ice cream 🍨. As the crime rate increases, ice cream sales also increase. So, one might suggest that criminals cause people to buy ice cream or that purchasing ice cream causes people to commit crimes. However, both are extremely unlikely.
Image Courtesy of Pinterest.
When taking a look at this graphic above, smoking becomes a very clear confounding variable when thinking about what causes many chronic illnesses and cancer. It may look like coffee is directly causing pancreatic cancer, since you are focused on it and are measuring its effects, but smoking contributes as well, without you realizing. That rings the question: how accurate is it that exposure to coffee leads to pancreatic cancer or that the two are even correlated? This is why are so important!
Here is a quick reminder of the different research methods and their strengths/weaknesses:
| Research Method | Basic Purpose | How Conducted | Manipulated | Strengths | Weaknesses |
| Descriptive | To observe and record behavior | Do , , or | Nothing | require only one participant; may be done when it is not ethical to manipulate variables; may be done quickly and inexpensively (compared with experiments) | Uncontrolled variables mean cannot be determined; single cases may be misleading |
| Correlational | To detect naturally occurring relationships; to assess how well one variable predicts another | Collect data on two or more variables; no manipulations | Nothing | Works with large groups of data, and may be used in situations where an experiment would not be ethical or possible | Does not specify |
| Experimental | To explore | Manipulate one or more variables; use random assignment | The independent variable(s) | Specifies , and variables are controlled | Sometimes not feasible; results may not generalize to other contexts; not ethical to manipulate certain variables |
Table Courtesy of Myers AP Psychology Textbook - 2nd Edition
Practice AP Question
The following prompt is from the 2013 AP Psychology Exam (#1).
In response to declining reading scores in local schools, John wrote an editorial suggesting that schools need to increase interest in reading books by providing students with incentives. Based on research showing a relation between use of incentives and student reading, he recommended providing a free pizza coupon for every ten books a student reads. 🍕
This question had several parts, but let's relate it to research methods.
Could we trust his argument? Does he have the right evidence?
Not really, right? He never ran an experiment and actually implied causation. That is something we could never do. Association does not equal causation.
Just because he found a "relation" between incentives and student reading, it doesn't mean that using incentives could increase student reading. Therefore, we could easily refute John's argument and prove that providing an incentive would not help increase interest in reading books. 📚
Key Terms to Review (12)
Case Studies
: Case studies are detailed investigations into one individual or small group. These studies provide rich qualitative data and can offer insights into unique situations that may not be captured in larger-scale research methods.Cause and Effect
: Cause and effect is a relationship between events or things, where one is the result of the other. This is a combination of action and reaction.Confounding Variables
: These are factors other than the independent variable that may cause a result in an experiment, making it difficult to determine if the results were due to the independent variable or these other factors.Correlation
: Correlation refers to the statistical relationship between two variables. If one variable changes, there tends to be a consistent change in the other variable too.Correlational Research Method
: Correlational research methods involve measuring two variables and analyzing how they relate to each other - whether they move together (positive correlation), move in opposite directions (negative correlation), or have no relationship (zero correlation).Descriptive Research Method
: This method involves observing and describing behavior without manipulating variables. It often involves naturalistic observations, surveys, and case studies.External Validity
: External validity refers to how well the results of a study can be generalized or applied to other situations and people.Internal Validity
: Internal validity refers to how well an experiment is done, especially whether it avoids confounding (more than one possible independent variable [cause] acting at the same time).Naturalistic Observations
: Naturalistic observation is an observational method where researchers observe individuals' behavior in their natural environment without intervention or manipulation.Research Question
: A research question is a clear, focused, concise, complex and arguable question around which you center your research. It's the question to which you want to find an answer through your study.Surveys
: Surveys are a research method that involves asking a large number of people questions about their thoughts, feelings, behaviors, or attitudes.Validity
: Validity refers to the extent to which a test or experiment measures what it is supposed to measure.1.4 Selecting a Research Method
4 min read•january 6, 2023
Sadiyya Holsey
Sadiyya Holsey
When selecting a research method, the researcher should write down their goals and what their is.
Selecting a research method depends on what a researcher wants to show or prove. For example, if a researcher only wants to show a between two variables, then an experiment would not be optimal. If a researcher wants to show between two variables, then an experiment is the best method.
When performing any research, the researcher should be sure of their results to ensure that they show true relationships. ✔️
Validity
Selecting a research method also depends on the of the experiments. There are two types of : and .
refers to how generalizable the results of the experiment are. For example, if the study on a drug is done on an Asian, middle-aged, average-weight man with high blood pressure, can the results be generalized to the population?
is when a study shows a truthful cause-and-effect relationship and the researcher is confident that the changes in the dependent variable were produced only by the independent variable. A confounding variable hurts the because it creates lower confidence in the research conclusion.
Confounding Variables
limit the confidence that researchers have in their conclusions. To recall from the last key topic, the confounding variable is an outside influence (variable) that changes the effect of a dependent and independent variable. For example, we looked at the between crime ⛓️ and the sale of ice cream 🍨. As the crime rate increases, ice cream sales also increase. So, one might suggest that criminals cause people to buy ice cream or that purchasing ice cream causes people to commit crimes. However, both are extremely unlikely.
Image Courtesy of Pinterest.
When taking a look at this graphic above, smoking becomes a very clear confounding variable when thinking about what causes many chronic illnesses and cancer. It may look like coffee is directly causing pancreatic cancer, since you are focused on it and are measuring its effects, but smoking contributes as well, without you realizing. That rings the question: how accurate is it that exposure to coffee leads to pancreatic cancer or that the two are even correlated? This is why are so important!
Here is a quick reminder of the different research methods and their strengths/weaknesses:
| Research Method | Basic Purpose | How Conducted | Manipulated | Strengths | Weaknesses |
| Descriptive | To observe and record behavior | Do , , or | Nothing | require only one participant; may be done when it is not ethical to manipulate variables; may be done quickly and inexpensively (compared with experiments) | Uncontrolled variables mean cannot be determined; single cases may be misleading |
| Correlational | To detect naturally occurring relationships; to assess how well one variable predicts another | Collect data on two or more variables; no manipulations | Nothing | Works with large groups of data, and may be used in situations where an experiment would not be ethical or possible | Does not specify |
| Experimental | To explore | Manipulate one or more variables; use random assignment | The independent variable(s) | Specifies , and variables are controlled | Sometimes not feasible; results may not generalize to other contexts; not ethical to manipulate certain variables |
Table Courtesy of Myers AP Psychology Textbook - 2nd Edition
Practice AP Question
The following prompt is from the 2013 AP Psychology Exam (#1).
In response to declining reading scores in local schools, John wrote an editorial suggesting that schools need to increase interest in reading books by providing students with incentives. Based on research showing a relation between use of incentives and student reading, he recommended providing a free pizza coupon for every ten books a student reads. 🍕
This question had several parts, but let's relate it to research methods.
Could we trust his argument? Does he have the right evidence?
Not really, right? He never ran an experiment and actually implied causation. That is something we could never do. Association does not equal causation.
Just because he found a "relation" between incentives and student reading, it doesn't mean that using incentives could increase student reading. Therefore, we could easily refute John's argument and prove that providing an incentive would not help increase interest in reading books. 📚
Key Terms to Review (12)
Case Studies
: Case studies are detailed investigations into one individual or small group. These studies provide rich qualitative data and can offer insights into unique situations that may not be captured in larger-scale research methods.Cause and Effect
: Cause and effect is a relationship between events or things, where one is the result of the other. This is a combination of action and reaction.Confounding Variables
: These are factors other than the independent variable that may cause a result in an experiment, making it difficult to determine if the results were due to the independent variable or these other factors.Correlation
: Correlation refers to the statistical relationship between two variables. If one variable changes, there tends to be a consistent change in the other variable too.Correlational Research Method
: Correlational research methods involve measuring two variables and analyzing how they relate to each other - whether they move together (positive correlation), move in opposite directions (negative correlation), or have no relationship (zero correlation).Descriptive Research Method
: This method involves observing and describing behavior without manipulating variables. It often involves naturalistic observations, surveys, and case studies.External Validity
: External validity refers to how well the results of a study can be generalized or applied to other situations and people.Internal Validity
: Internal validity refers to how well an experiment is done, especially whether it avoids confounding (more than one possible independent variable [cause] acting at the same time).Naturalistic Observations
: Naturalistic observation is an observational method where researchers observe individuals' behavior in their natural environment without intervention or manipulation.Research Question
: A research question is a clear, focused, concise, complex and arguable question around which you center your research. It's the question to which you want to find an answer through your study.Surveys
: Surveys are a research method that involves asking a large number of people questions about their thoughts, feelings, behaviors, or attitudes.Validity
: Validity refers to the extent to which a test or experiment measures what it is supposed to measure.
Resources
© 2024 Fiveable Inc. All rights reserved.
AP® and SAT® are trademarks registered by the College Board, which is not affiliated with, and does not endorse this website.