• Descriptive methods
  • Naturalistic observation
  • Survey methods
  • case studies
  • Archival research
• Experimental methods
  • Constants and variables
  • Experimental control
  • Experimental designs

In many ways, psychology's goals are similar to the goals of the physical sciences, but we're concerned with the behavior of people (and animals) rather than the physical world.
• Description of behavior describe events, what changes what might affect change, what might be related twhat, etc.
• Prediction of behavior given X what will likely happen
• Control of behavior for intervention kinds of purposes (related tprediction of behavior)
• Causes of behavior sometimes predictions aren't enough, want tknow how the X and the outcome are related
• Explanation of behavior a complete theory of the how's and why's

One big difference is that human behavior (and animals) is typically much more variable than most physical systems.
To address this:
• we use a lot of statistical procedures.
• we reduce variability by using various methods of control.

General Approaches to Research Methodologies

• Descriptive (Correlational or Observational) : Systematic observation of naturally occurring events
• Experimental: Exhaustive observation of nonnaturally occurring events (controlled laboratory experiments)

  The emphasis in the course will be on survey and experimental designs.

• Advantages
  • may see patterns of behaviors that are very complex and realized on in particular settings
  • often very useful when little is known about the subject of study
  • may learn about something that never would have thought of looking at in an experiment
• Disadvantages
  • Causality is a problem
  • internal validity problems due to lack of control
    • Every confound is a threat, Lots of alternative explanations
      • Stork example
    • Directionality of the relationship isn't known
      • Airplane turbulents and coffee spills
      • Happy people tend to sleep well (or is it that you sleep well when you're happy?)
    • Sometimes the results are not reproducible

• Naturalistic observation
  • a scientific approach that focuses on observation and description within its natural setting; it is generally used with naturally occurring and ongoing phenomena with little experimenter intervention
  • Often a first step in the research project, helps to identify what some of the important variables are, then the next step is to move into more controlled settings for further tests
• Survey methods
  • the technique of obtaining a limited amount of information (e.g. asking people about their opinions or behaviors) from a large number of people, usually through random sampling
  • widely used methodology
• Case histories and Clinical methodologies
  • Intensive study of a single person, a very traditional method (e.g., Freud)
• Archival research
  • Using data sets that have been collected by other researchers (could be quantitative or qualitative depending on the nature of the data)

The Experimental Method

• The researcher systematically manipulates the environment to observe the effect of this manipulation on behavior.
  (manipulates the independent variable[s] and controls all other relevant variables either by randomization or by direct experimental control)
• The method allows a causal inference tbe made: Any change in the dependent variable was caused by the manipulation of the independent variable

Constants vs. variables - characteristics of the psychological situations

• Constants - have the same value for all individuals in the situation
• Variables - have potentially different values for each individual in the situation

• Conceptual vs. Operational

  • Conceptual variables are often abstract theoretical entities which have corresponding operational variables in the experiment (concrete so that they can be measured or manipulated)

  • Within each of these two types exist the following subclasses of variables

    • Independent variables - these are the variables that are manipulated by the experimenter, results in the different conditions in an experiment
      • Event/Stimulus manipulations - manipulate characteristics of the stimuli, context, etc.
      • Instructional manipulations - different groups are given different instructions
      • Subject manipulations - there are (pre-existing mostly) differences between the subjects in the different conditions

    • Dependent variables - these are the variables that are measured by the experimenter, they are "dependent" on the independent variables (if there is a relationship between the IV and DV as the hypothesis predicts).

    • Control (extraneous) variables
      Holding things constant - Controls for excessive random variability

    • Random variables - may freely vary, to spread variability equally across all experimental conditions
      • Randomization
        • A procedures that assure that each level of an extraneous variable has an equal chance of occurring in all conditions of observation.
        • If this is true, then the most likely outcome is that the extraneously variable is, on average, equally represented in all conditions.
        • On average, the extraneous variable is not confounded with our manipulated variable.

    • Confound variables
      • Other variables, that haven't been accounted for (manipulated, measured, randomized, controlled) that co-vary with the IV and can impact changes in the dependent variable(s)

Experimental Control -

Why do we need to do it?
1) We want to minimize excessive variability.
2) To reduce the potential of confoundings.
• our goal is to test the possibility of a relationship between the variability in our IV and how that affects our DV. However, if there are other variables that influence our DV, how do we know that the observed differences are due to our IV and not some other variable

  Sources of Total Variability:

  I. Nonrandom (NR) Variability - systematic variation

  A. (NR_exp)manipulated independent variables (IV)
  i. our hypothesis is that changes in the IV will result in changes in the DV
  B. (NR_other)extraneous variables (EV) which covary with IV
  i. other variables that also vary along with the changes in the IV, which may in turn influence changes in the DV

  II. Random (R) Variability

  A. imprecision in manipulation (IV) and/or measurement (DV)
  B. randomly varying extraneous variables (EV)

  Sources of Total (T) Variability: T = NR_exp + NR_other +R

  Our goal is to reduce R and NR_other so that we can detect NR_exp. That is, so we can see the changes in the DV that are due to the changes in the independent variable(s).

  Many of the statistics that we'll use in this course are ratios that compare variability due to chance with total variability (which is usually variability due to chance plus variablility due to a treatment). As an analogy, consider the ratio as a scale. Sources of variability are weights put on the scale. Large weights equal large sources of variability, small weights are small sources. "Significant" effects are those where the "tilt" measured by the scale is "big" enough to see (easily).
  	Here we have a pretty large treatment effect and relatively small random and non-random extraneous variability. So the treatment weight has a big effect on the tilt.
  	Here we have a pretty small treatment effect and some random variability and a large amount of non-random extraneous variability. So the treatment weight has a relatively weak effect on the tilt.
  Try this at home: This analogy actually works better if you use your hands rather than a real scale. Get a bunch of small pairs of objects (different sized batteries work really nicely). Assign each set of weights to the different sources of variability. Then close your eyes and have somebody put different combinations of the variability in your hands. Which hand is heavier? When is this distinction easy to make, when is it hard?

Comparison

• An experiment always makes a comparison, so it must have at least two groups
  • Sometimes there are baseline, or control groups
  • Without control groups if is harder to see what is really happening in the experiment, it is easier to be swayed by plausibility or inappropriate comparisons

Within vs. Between group designs
• Some experimental designs use participants as their own control group. A within (repeated-measures) variable is one in which the participant responds in all levels of the independent variable. This reduces overall variability by removing variance due to individual differences.

Statistical control

• Another way to reduce variability is to partition the variance and remove portions of it in systematic ways.

So how does this all relate to this course on statistics?? Some common experimental designs.

At the heart of an experiment is a comparison between two conditions. In other words, you (the experimenter) will always be comparing at least two things. This may include comparing your sample with a known population, or two (or more) different samples against each other, or even multiple scores within a single sample.

 Below is a decision tree that lists some common experimental designs and the kind of statistical analysis test that is used to analyze each one. Later in the course we'll be learning about how to conduct (some of) these statistical tests (that's what the "lab" buttons by the different tests are for).

Click the button below. This will open another window in which there is a decsion tree. This tree will help you decide, based on your experimental design, which statistical test to use.

• Data that need to be considered on a pair-by-pair basis (e.g., repeated measure, correlation, regression, etc.) need data in multiple columns so that SPSS knows which data to pair up. Otherwise, SPSS considers the data one-by-one and only needs to "know" which condition the data are in.
• Between vs. within
  • if your Independent variable is manipulated between groups, then your datafile should have one column that specifies the different levels of that variable (so it basically is telling SPSS what condition that particular row of data corresponds to).
  • if your independent variable is manipulated within groups, then you need a separate column of data for each level of that variable.
• Correlation and regression - each of the two (or more) variables need to be given their own column