Chapter Two: Psychology research

Chapter Two: Psychology research

1. Steps in a scientific investigation

   • 1. Formulate a Testable Hypothesis
     • A hypothesis is a tentative statement or prediction about the relationship between two or more variables.
       • Variables are any changeable condition that are controlled or observed in a study.
     • Operational definition: describes the actions or operations that will be used to control or measure a variable.
   • 2. Select the Research Method and Design the Study
     • What research method is most practical? – e.g. experiment, survey, case study, naturalistic observation. Decide when to conduct survey, how many people, where will researchers get the participants
       • Subjects or Participants - persons or animals whose behaviour is systematically observed.
   • 3. Collect the Data
     • Data collection techniques - are procedures for making empirical observations and measurements, e.g. direct observation, questionnaires, interviews, psychological tests, physiological recordings, examination of archival records.
   • 4. Analyze the Data and Draw Conclusions
     • Observations are usually converted into numbers – raw data Researchers use statistics to analyze their data and decide whether or not their hypothesis is supported.
   • 5. Report the Findings
     • Scientists make a concise summary of the study and the findings. Often the report is presented at a scientific meeting and submitted to a journal for publication. This allows other experts to evaluate and critique the findings. It weeds out erroneous findings.
   • Advantages of Scientific Approach
     • 1. Clarity and precision are used to formulate hypotheses and variables. Therefore, it allows us to communicate better about portant ideas.
     • 2. Relative intolerance of error - yields more accurate and dependable information than just common sense or traditional approaches.

2. EXPERIMENTAL RESEARCH

   • This method is useful to allow researchers to detect cause-and-effect relationships. Most commonly used by psychologists. In experiments psychologists want to know how one variable affects another variable, particularly whether changes in one variable cause changes in another variable.
     • Independent variable is a condition or event that the experimenter varies in order to see its impact on another. The experimenter manipulates or controls this factor by putting the subjects into two different groups or conditions. This variable is believed to effect or cause the second variable known as the dependent variable which is measured.
       • • Dependent variable (DV): Variable the experimenter measures, after making changes to the IV that are assumed to affect the DV.
       • • Independent variable (IV): Variable the experimenter manipulates (i.e. changes) – which is assumed (hypothesized) to have a direct effect on the dependent variable.
   • Experimental and Control Groups In an experiment an experimenter usually works with two groups of similar subjects who are treated differently only with regard to the independent variable.
     • Experimental group consists of the subjects who receive some special treatment in regard to the independent variable.
     • Control group consists of similar subjects who do not receive the special treatment given to the experimental group. This allows researchers to learn the effect of the IV on the DV.
       • Extraneous variables are any variables other than the independent variable that seem likely to influence the dependent variable in a specific study. Experimenters try to make sure that the experimental and control groups are the same except for the independent variable.
         • One way to control for extraneous variable is to assign subjects to the experimental group or the control group randomly. Random assignment of subjects occurs when all subjects have an equal chance of being assigned to any group or condition in the study.
       • Confounding of variables - occurs when two variables are linked together in a way that makes it difficult to sort out their specific effects. When an extraneous variable is confounded with an independent variable, the researcher cannot tell which one is having the effect.
   • Variations in Designing Experiments
     • 1. Sometimes experimenters find it advantageous to only use one group of subjects and they serve as their own control group. A single group is exposed to an experimental condition and a control condition
       • 2. It is possible to manipulate more than one independent variable. This variation is often used to see if there is an interaction between the two independent variables --- does the one variable actually depend on the other.
         • 3. You can also have more than one dependent variable in a single study.
   • Advantages and Disadvantages of Experimental Research
     • Main Advantage: Experiments are done under very controlled conditions, therefore, it permits conclusions about cause and effect relationships
       • Disadvantages- 1. Experiments are often artificial. Researchers often construct simple, contrived situations because of the degree of control needed to do an experiment. When experiments are very artificial, it makes one wonder if the findings actually apply to everyday behaviour outside the experimental lab.
         • 2. Some hypotheses do not lend themselves to manipulating independent variables due to ethical or practical reasons.

3. DESCRIPTIVE/CORRELATIONAL RESEARCH

   • When research does not lend itself to using the experimental method we can use descriptive or correlational research methods. These types of studies cannot demonstrate cause and effect because of the lack of control. Descriptive/correlational methods only permit investigators to describe patterns of behaviour and discover associations between variables. But it is still very valuable information. These methods include naturalistic observation, case studies and surveys.
     • Naturalistic observation: a researcher engages in careful observation of behaviour without intervening directly with the subjects. Examples include using a one way mirror in a day care, videotaping people (when ethical) and observing people without them knowing. Advantage: Less artificial than the lab Disadvantage: Trouble being unobtrusive so that experimenter does not affect the subject’s behaviour.
     • Various data collection techniques are used in case studies - interviews, direct observation of the subject, examination or records, and psychological testing. Psychological autopsy refers to studying a person after he/she is deceased.
     • Surveys: In a survey researchers use questionnaires or interviews to gather information about specific aspects of the subjects’ behaviour. Surveys are used to obtain information on aspects of behaviour that are difficult to observe directly. E.g. sexual behaviour, attitudes, opinions Useful for getting information from a large number of people. Unfortunately, they depend on self-report data. Sometimes the subject might write something that is not entirely true, or they forget, or are engaging in wishful thinking when they answer the questions.
   • Advantages and Disadvantages of Descriptive/Correlational Research Advantage - Researchers can explore questions that could not be examined with experimental procedures. It broadens the scope of phenomena that psychologist area able to study. Disadvantage - Cannot demonstrate conclusively that two variables are causally related (No cause and effect)
   • Descriptive Statistics
     • Central Tendency – Measures of central tendency answer the question, “Where do most of your results fall?”
       • median – This is the best measure to use if one or a few of the scores are quite different than most. This is calculated by arranging all the numbers in numerical order and finding the centre figure. A common use of the median is to report the middle income level in a Country or community. Using the Median means that the middle income is not distorted by those people who earn very little or those people who earn a lot.
     • mean or average – most commonly used in psychological experiments. (The average weight of a student in grade 7 is 95 lbs.) Add up the value of each measurement and divide by the number of measurement (or observations).
     • Mode – number that occurs the most frequently in the data
   • Measures of Variability
     • Usually when you do an experiment you end up with two sets of numbers, one for the control group and one for the experimental group. We use a statistic to describe the degree of difference between the two sets of numbers. The two means could be exactly the same yet the variability in the sets of numbers may be quite different. Standard Deviation – The more different the numbers are from each number the larger the standard deviation.
   • Correlation - In correlational studies you want to know how strong of an association there is between two variables, both of which are measured. E.g. High entrance exam scores are associated or correlated with high university averages and low scores with low averages in university.
     • Correlation coefficient- Scatterplots are useful but we need to be able to use a number to indicate the degree of relationship between 2 sets of scores. A correlation coefficient does this. Correlation coefficients can be any number between 0 and plus or minus 1 0 means no correlation. If you know one score, you know nothing about the second score. Plus or minus one is a perfect correlation. If you know the one score, you can be sure of the score on the second. Usually it is somewhere between 0 and -1 or +1. The closer to 0 the weaker the correlation. Positive correlation – high values on one measure are related to high values on the other measure and low values are related to low values. (More hours studying related to higher grades) Negative correlation – high values on one measure are related to low values on the other measure. (More alcohol on weeknights related to lower grades) The + or – sign indicates whether the correlation is positive or negative The number indicates t

4. Extra knowledge

   • PLACEBO EFFECTS
     • Placebo – something that appears like medication but actually has no effect on the body as it is fake (sugar pill/vitamin). Placebos are used in studying effects of medication because people may expect a change in their feelings or behaviour because they have taken “medication”. Placebo effects – refers to a change that occurs because people expect a change
   • DISTORTIONS IN SELF-REPORT DATA
     • Self-report data – information given by the subject about the subject. Self-report data is often used and may be useful but there can be some problems. Social desirability bias – tendency to give socially approved answers when questioned about oneself. Implicit Measures can provide self-report data that bypasses one’s consciousness.
   • EXPERIMENTER BIAS
     • Researchers have certain expectations of the outcome of their hypotheses. Experimenter bias occurs when researcher’s expectations or preferences about the outcome of the study influence the results obtained. See Rosenthal’s study. One method used to prevent experimenter bias, is called the double-blind procedure. In this case neither the subjects nor the experimenter is aware of who is in the experimental and who is in the control group. Subjects could be given placebos that look like real medication or simply not informed. Researchers may have their graduate students deliver the meds and placebos to the subjects and not tell them which is which.
   • INTERNET AND PSYCHOLOGICAL RESEARCH
     • Internet-mediated research – studies in which data is collected using the Web. Any type of research in this chapter can be done through the web with surveys and experiments being the most common. One benefit is that researchers can get larger samples through the internet and often more diverse samples. Special populations can be located more easily through the internet than through traditional methods. As well, once it is set up, it takes less effort on the part of the researcher. Another benefit, is that people can report their feelings, thoughts and behaviour privately and are less likely to be use social desirable responses. E.g. marital issues Disadvantages include sampling bias since not everyone has access to the internet, and those who do are likely younger, smarter and able to afford a computer. Only a few of the people who see a research opportunity online actually participate. Therefore they may be somewhat different from the vast majority who do not. Also there is
   • Anecdotal Evidence - Personal stories about specific incidents and experiences. They are easily believed because they are often vivid and easily remembered. Anecdotes (stories) are not necessarily representative of the population. They are much like a single case study or have a sample size of one (n=1). Also, anecdotes consist of self-report data and can be distorted due to the social desirablilty bias (wanting others to think positively of them).

Images and files in the map