Chapter 2 – Research Methods
A – How do Psychologists Develop New Knowledge?
- Psychologists, like researchers in all other sciences, use the scientific method to test their ideas empirically.
- The scientific method is a five-step process for logical investigation of a hypothesis under conditions designed to
control biases and subjective judgments.
- At the heart of this testing procedure is the empirical investigation, an approach to research that relies on
sensory experience and observation as research data. Empirical means experience based-as contrasted with
speculation based solely on faith, hope, authority, or common sense.
- Theory is a testable explanation for a set of facts or observations. In science, a theory is not just a speculation or a
guess. Examples of well-supported theories include Einstein's theory of relativity, the germ theory of disease,
Darwin's theory of natural selection, and, in psychology, social learning theory.
- Emily Rosa of Lockland, CO challenged a widely held belief in the power of therapeutic touch (TT). In the early
1990s, TT was touted as a medical theory, and Emily's mom, a nurse, had explained to her how TT practitioners
attempted to promote healing by moving their hands over the patient's body without directly touching it. They
believed that they were detecting and manipulating energy did radiating from the body. They claimed it treated
medical and psychological problems. Emily Rosa suspected that TT practitioners were really detecting their own
beliefs and expectations, rather than a 'human energy field'. She used the scientific method to put their claims to
into a simple experimental test.
B – The Five Steps of the Scientific Method
- These steps are essentially the same whether the study involves psychology, biology, chemistry, astronomy, or any
other scientific discipline.
Developing a Hypothesis
- Hypothesis is a statement predicting the outcome of a scientific study; a statement describing the relationship
among variables in a study.
- The term literally means “little theory” because it often represents only one piece of a larger theoretical puzzle.
- Operational definitions are specific descriptions of concepts involving the conditions of a scientific study.
Operational definitions are stated in terms of how the concepts are to be measured or what operations are being
employed to produce them.
- Emily Rosa had to specify the exact procedures she would use in setting up the experimental conditions and
measuring the results.
Performing a Controlled Test
- Emily invited each of 21 TT practitioners (1-27 years of experience) to determine which of their two hands was
closest to one of her own hands.
- An independent variable is a stimulus condition so named because the experimenter changes it independently of
all the other carefully controlled experimental conditions.
- In Emily’s experiment, randomization was achieved by a coin flip, which determined whether she presented her
hand above the practitioner’s left or right hand.
- Random presentation is a process by which chance alone determines the order in which the stimulus is
presented.
Gathering Objective Data
- Data is pieces of information, especially information gathered by a researched to be used in testing a hypothesis.
- In Emily’s experiment, the data consisted of the number of correct and incorrect responses to the placement of
her hand. Such responses are referred to as the dependent variable.
- Dependent variable is the measured outcome of a study, the responses of the subjects in a study. Experimenters
must give an operational definition, must specify the procedures that were used in measuring the responses
being observed.
Analyzing the Results and Accepting or Rejecting the Hypothesis
- The research must examine the results to see whether the hypothesis survived the test. Statistical analysis can tel
the research whether the observed results rise to the level of significance—that is, whether the results are likely
due to the independent variable or merely due to chance.
, - In Emily’s experiment, the statistical analysis was remarkably simply. The chances of getting a correct answer
merely by guessing were 50%. They had no ability to sense the human energy field.
Publishing, Criticizing, and Replicating the Results
- Emily published her results in the Journal of the American Medical Association.
- Replicate refers to doing a study over to see whether the same results are obtained. As a control for bias,
replication is often done by someone other than the research who performed the original study.
- No one replicated Emily’s experiment. Fewer than 2% of the papers submitted to psychological journals get into
print without major revisions. In fact, the majority never see print at all.
C – Types of Psychological Research
- Not all research is created equal. We look at two methodologies: the experiment, and a variety of quasi-
experimental methods.
Experimental Method
- An experiment is a kind of research in which the researcher controls all the conditions and directly manipulates
the conditions, including the independent variable.
- Confounding/extraneous variables are variables that have an unwanted influence on the outcome of an
experiment.
- Controls are constraints that the experimenter places on the experiment to ensure that each subject has the
exact same conditions.
- Random assignment is when each subject of the sample has an equal likelihood of being chosen for the
experimental group of an experiment.
Non-experimental Methods
- Some situations, due to ethical or practical reasons, we can’t always do an experiment.
- We can’t give people lung cancer and then try to cure it. We choose subjects based on the preexisting conditions
of having lung cancer and try to cure them.
- Ex post facto is research in which we choose subjects based on a pre-existing condition.
Correlational Studies
- Correlational study is a type of research that is mainly statistical in nature. Correlational studies determine the
relationship/correlational between two variables.
- Example: saying smoking and lung cancer were related. However, correlation doesn’t necessarily mean causation.
- Scientist usually expresses the degree of correlation as a number. This requires calculating a statistic known as th
correlation coefficient, often symbolized in formulas as the letter r.
- If the variables show a zero correlation, their correlation is zero. If two variables show a relationship in which they
vary in the same direction (either increasing or decreasing), their correlation is positive. If two variables show a
relationship in which they vary in different directions (one increase and one decrease), their correlation is
negative.
- A survey is a quasi-experiment method in which questions are asked to subjects. When designing a survey, the
researcher has to be careful that the questions are not skewed or biased toward a particular answer.
- Naturalistic observation is a research method in which subjects are observed in their natural environment. This
method is good for studies including child-rearing, people’s shopping habits, or public courting behaviors. This
ensures that the behavior being observed is the actual behavior in its natural state. However, naturalistic
observations are made under far less controlled conditions than experiments.
- Longitudinal study is a type of study in which one group of subjects is followed and observed (examined,
surveyed, ect.) for an extended period of time (years). The benefits of this research are that you have the same
subject group throughout but cost time and expense.
- To avoid the time and expense, cross-sectional study is a study in which a representative cross section of the
population is tested or surveyed at one specific time but it’s not so accurate.
- Yielding better results, cohort-sequential study is a research method in which a cross section of the population is
chosen and then each cohort/group is followed for a short period of time,. It takes less time than the longitudinal
design, is much less susceptible to bias, and therefore yields more accurate data than a cross-sectional study.
D – Sources of Bias in Research (or Anywhere Else)
- Personal bias is when the research allows personal beliefs to affect the outcome of a study.
A – How do Psychologists Develop New Knowledge?
- Psychologists, like researchers in all other sciences, use the scientific method to test their ideas empirically.
- The scientific method is a five-step process for logical investigation of a hypothesis under conditions designed to
control biases and subjective judgments.
- At the heart of this testing procedure is the empirical investigation, an approach to research that relies on
sensory experience and observation as research data. Empirical means experience based-as contrasted with
speculation based solely on faith, hope, authority, or common sense.
- Theory is a testable explanation for a set of facts or observations. In science, a theory is not just a speculation or a
guess. Examples of well-supported theories include Einstein's theory of relativity, the germ theory of disease,
Darwin's theory of natural selection, and, in psychology, social learning theory.
- Emily Rosa of Lockland, CO challenged a widely held belief in the power of therapeutic touch (TT). In the early
1990s, TT was touted as a medical theory, and Emily's mom, a nurse, had explained to her how TT practitioners
attempted to promote healing by moving their hands over the patient's body without directly touching it. They
believed that they were detecting and manipulating energy did radiating from the body. They claimed it treated
medical and psychological problems. Emily Rosa suspected that TT practitioners were really detecting their own
beliefs and expectations, rather than a 'human energy field'. She used the scientific method to put their claims to
into a simple experimental test.
B – The Five Steps of the Scientific Method
- These steps are essentially the same whether the study involves psychology, biology, chemistry, astronomy, or any
other scientific discipline.
Developing a Hypothesis
- Hypothesis is a statement predicting the outcome of a scientific study; a statement describing the relationship
among variables in a study.
- The term literally means “little theory” because it often represents only one piece of a larger theoretical puzzle.
- Operational definitions are specific descriptions of concepts involving the conditions of a scientific study.
Operational definitions are stated in terms of how the concepts are to be measured or what operations are being
employed to produce them.
- Emily Rosa had to specify the exact procedures she would use in setting up the experimental conditions and
measuring the results.
Performing a Controlled Test
- Emily invited each of 21 TT practitioners (1-27 years of experience) to determine which of their two hands was
closest to one of her own hands.
- An independent variable is a stimulus condition so named because the experimenter changes it independently of
all the other carefully controlled experimental conditions.
- In Emily’s experiment, randomization was achieved by a coin flip, which determined whether she presented her
hand above the practitioner’s left or right hand.
- Random presentation is a process by which chance alone determines the order in which the stimulus is
presented.
Gathering Objective Data
- Data is pieces of information, especially information gathered by a researched to be used in testing a hypothesis.
- In Emily’s experiment, the data consisted of the number of correct and incorrect responses to the placement of
her hand. Such responses are referred to as the dependent variable.
- Dependent variable is the measured outcome of a study, the responses of the subjects in a study. Experimenters
must give an operational definition, must specify the procedures that were used in measuring the responses
being observed.
Analyzing the Results and Accepting or Rejecting the Hypothesis
- The research must examine the results to see whether the hypothesis survived the test. Statistical analysis can tel
the research whether the observed results rise to the level of significance—that is, whether the results are likely
due to the independent variable or merely due to chance.
, - In Emily’s experiment, the statistical analysis was remarkably simply. The chances of getting a correct answer
merely by guessing were 50%. They had no ability to sense the human energy field.
Publishing, Criticizing, and Replicating the Results
- Emily published her results in the Journal of the American Medical Association.
- Replicate refers to doing a study over to see whether the same results are obtained. As a control for bias,
replication is often done by someone other than the research who performed the original study.
- No one replicated Emily’s experiment. Fewer than 2% of the papers submitted to psychological journals get into
print without major revisions. In fact, the majority never see print at all.
C – Types of Psychological Research
- Not all research is created equal. We look at two methodologies: the experiment, and a variety of quasi-
experimental methods.
Experimental Method
- An experiment is a kind of research in which the researcher controls all the conditions and directly manipulates
the conditions, including the independent variable.
- Confounding/extraneous variables are variables that have an unwanted influence on the outcome of an
experiment.
- Controls are constraints that the experimenter places on the experiment to ensure that each subject has the
exact same conditions.
- Random assignment is when each subject of the sample has an equal likelihood of being chosen for the
experimental group of an experiment.
Non-experimental Methods
- Some situations, due to ethical or practical reasons, we can’t always do an experiment.
- We can’t give people lung cancer and then try to cure it. We choose subjects based on the preexisting conditions
of having lung cancer and try to cure them.
- Ex post facto is research in which we choose subjects based on a pre-existing condition.
Correlational Studies
- Correlational study is a type of research that is mainly statistical in nature. Correlational studies determine the
relationship/correlational between two variables.
- Example: saying smoking and lung cancer were related. However, correlation doesn’t necessarily mean causation.
- Scientist usually expresses the degree of correlation as a number. This requires calculating a statistic known as th
correlation coefficient, often symbolized in formulas as the letter r.
- If the variables show a zero correlation, their correlation is zero. If two variables show a relationship in which they
vary in the same direction (either increasing or decreasing), their correlation is positive. If two variables show a
relationship in which they vary in different directions (one increase and one decrease), their correlation is
negative.
- A survey is a quasi-experiment method in which questions are asked to subjects. When designing a survey, the
researcher has to be careful that the questions are not skewed or biased toward a particular answer.
- Naturalistic observation is a research method in which subjects are observed in their natural environment. This
method is good for studies including child-rearing, people’s shopping habits, or public courting behaviors. This
ensures that the behavior being observed is the actual behavior in its natural state. However, naturalistic
observations are made under far less controlled conditions than experiments.
- Longitudinal study is a type of study in which one group of subjects is followed and observed (examined,
surveyed, ect.) for an extended period of time (years). The benefits of this research are that you have the same
subject group throughout but cost time and expense.
- To avoid the time and expense, cross-sectional study is a study in which a representative cross section of the
population is tested or surveyed at one specific time but it’s not so accurate.
- Yielding better results, cohort-sequential study is a research method in which a cross section of the population is
chosen and then each cohort/group is followed for a short period of time,. It takes less time than the longitudinal
design, is much less susceptible to bias, and therefore yields more accurate data than a cross-sectional study.
D – Sources of Bias in Research (or Anywhere Else)
- Personal bias is when the research allows personal beliefs to affect the outcome of a study.
