- Define the principles of the scientific method
- Understand some of the challenges of conducting psychological research


[SLIDE 1]

Psychologists, like other scientists, use careful means to observe and measure behavior and the factors that influence behavior. Psychologists use the scientific method to decide what kind of evidence they need and how to pursue it. They also select the population they intend to study, such as U.S. teenagers or older adults, and try to obtain a representative sample of that population. 


[SLIDE 2]

The scientific method is an organized way of using experience and testing ideas to expand and refine knowledge. Psychologists do not necessarily follow the steps of the scientific method as we might follow a recipe in a cookbook, but research is guided by certain principles.

Psychologists usually begin by formulating a research question. Research questions can have many sources: our daily experiences, psychological theory, and even folklore all help generate questions for research. As another example, social cognitive principles of observational learning may prompt research on the effects of television violence. Research questions may also arise from common knowledge.


[SLIDE 3]

A research question may be studied as a question or reworded as a hypothesis. A hypothesis is a statement about behavior or mental processes that is testable through research. For example, a hypothesis about day care might be that preschoolers who are placed in day care will acquire greater social skills in relating to peers than preschoolers who are cared for in the home.


[SLIDE 4]

Psychologists next examine the research question or test the hypothesis through controlled methods such as an experiment. For example, we could take a group of preschoolers who attend day care and another group that does not and introduce each to a new child in a controlled setting such as a child-research center. We could then observe how children in each group interact with the new acquaintance.

Psychologists draw conclusions about their questions or the accuracy of their hypotheses on the basis of their observations or findings. When their observations do not bear out their hypotheses, they may modify the theories from which the hypotheses were derived. Research findings often suggest refinements to psychological theories and new avenues of research. In our research on day care, we might find that children in day care show greater social skills than children who are cared for in the home.


[SLIDE 5]

As psychologists draw conclusions from research evidence, they are guided by principles of critical thinking. For example, they try not to confuse correlations or associations between findings with cause and effect. For example, although more aggressive children apparently spend more time watching violent television shows, it may be erroneous to conclude from this kind of evidence that television violence causes aggressive behavior. 

A selection factor, a source of bias that may occur in research findings when participants are allowed to choose for themselves a certain treatment in a scientific study, may be at work because the children being studied choose for themselves what they will watch. Perhaps more aggressive children are more likely than less aggressive children to tune in to violent television shows.


[SLIDE 6]

To improve accuracy of research, psychologists carry out studies on a sample. The sample is a segment of a population. A population is a complete group of interest to researchers from which a sample is drawn. 

The sample must be drawn so that it accurately represents that population. Only representative samples allow us to generalize, or extend, our findings from research samples to target populations.


[SLIDE 7]

Many factors must be considered when interpreting the accuracy of the results of scientific research. One is the nature of the research sample.  


[SLIDE 8]

One way to achieve a representative sample is by means of random sampling. In a random sample, each member of a population has an equal chance of being selected to participate. Researchers can also use a stratified sample, which is selected so that identified subgroups in the population are represented proportionately in the sample.

For instance, 13% of the American population is African American. A stratified sample would thus be 13% African American. As a practical matter, a large randomly selected sample will show reasonably accurate stratification. A random sample of 1,500 people will represent the broad American population reasonably well. However, a sample of 20,000 European Americans or just men will not.


[SLIDE 9]

Large-scale magazine surveys of sexual behavior ask readers to fill out and return questionnaires. Although many thousands of readers complete the questionnaires and send them in, do the survey respondents represent the American population? Probably not. These and similar studies may be influenced by volunteer bias, which is a source of bias or error in research reflecting the prospect that people who offer to participate in research studies differ systematically from people who do not.

In the case of research on sexual behavior, volunteers may represent subgroups of the population, or of readers of the magazines in question, who are willing to disclose intimate information and therefore may also be likely to be more liberal in their sexual behavior. Volunteers may also be more interested in research than other people, as well as have more spare time.