CHAPTER SIX Conducting a Good Experiment I: Variables and Control 1 The Nature of Variables! Variable! A variable is an event or behavior that can assume at least two values.! Bridgman (1927) suggested that researchers should define their variables in terms of the operations needed to produce them. " Such definitions allow others to replicate your research and are called operational definitions. 2 Independent Variables! Independent Variables (IV s)! IV s are those variables that the experimenter purposely manipulates.! The IV constitutes the reason the research is being conducted; the experimenter is interested in determining what effect the IV has. 3 1
Types of Independent Variables! Physiological IV! The physiological state of the participant that the experimenter manipulates.! Experience IV! Manipulation of the amount or type of training or learning.! Stimulus or environmental IV! An aspect of the environment that the experimenter manipulates. 4 Participant Characteristics! Participant characteristics are aspects of the participant, such as age, sex or personality traits, that are treated as if they are IVs.! But they are not IV s because they cannot be manipulated by the experimenter. 5 Extraneous Variables (Confounders)! Extraneous variables! Uncontrolled variables that can cause unintended changes between groups.! Confounding! A situation in which the results of an experiment can be attributed to either the operation of an IV or to an extraneous variable.! When a study is confounded, there is no way to determine which variable is responsible for any observed differences.! Researchers learn from mistakes and control the extraneous variable in the next experiment. 6 2
Extraneous Variables (Confounders) A. Difference between two groups with no confounder operating. B. Difference between two groups when a confounder is present and has moved the groups closer together. C. Difference between two groups when a confounder is present and has moved the groups farther apart. 7 Dependent Variables! Dependent Variable (DV)! A response or behavior that is measured; the data or results of an experiment.! It is desired that any changes in the DV are directly related to the manipulation of the IV.! The literature review provides valuable guidelines as to which behavior or response should be measured.! Once the DV has been selected, method of measuring or recording it must be determined. 8 Recording or Measuring the DV! Correctness! Only the correct responses are counted.! Rate or Frequency! Rate of responding determines how rapidly responses are made during a specified time period.! The number of responses or events that occur within a specified time period is the frequency.! Degree or Amount! Latency of Duration 9 3
Different Rates of Responding. Recording More than One DV! If you have the measurement capabilities, there is nothing to prohibit the recording of more than one DV.! If recording an additional DV makes a meaningful contribution to your understanding of the phenomenon under study, then you should give it serious consideration. 11 Characteristics of a Good DV! A DV is valid when it measures what the experimental hypothesis says it should measure.! A good DV must be directly related to the IV and must measure the effects of the IV manipulation as the experimental hypothesis predicts it will.! A good DV is also reliable. 12 4
Nuisance Variables! Nuisance Variables! Unwanted variables that can cause the variability of scores within the experimental groups to increase.! Nuisance variables only increase the spread of scores within a distribution; they do not cause a distribution to change its location. 13 Nuisance Variables A. Spread of scores within a group when a nuisance variable is not operating. B. Spread of scores when a nuisance variable is operating. A. A comparison of two groups when a nuisance variable is not operating. B. A comparison of the same two groups when a nuisance variable is operating. 14 Controlling Extraneous Variables! The experimenter must exercise control over both extraneous variables and nuisance variables so the results of the experiment are as meaningful (no extraneous variables present) and clear (minimal influence of nuisance variables) as possible. 15 5
Basic Control Techniques! Randomization! A control technique that ensures that each participant has an equal chance of being assigned to any group in an experiment! Elimination! A control technique whereby extraneous variables are completely removed from an experiment.! Constancy! A control technique by which an extraneous variable is reduced to a single value that is experienced by all participants.! Balancing! A control procedure that achieves group equality by distributing extraneous variables equally to all groups.! Counterbalancing! A procedure for controlling order effects by presenting different treatment sequences. 16 Counterbalancing! Within-Subject counterbalancing! Presentation of different treatment sequences to the same participant. 17 Counterbalancing! Within-Group counterbalancing! Presentation of different treatment sequences to different participants. " Three basic requirements: " Each treatment must be presented to each participant an equal number of times. " Each treatment must occur an equal number of times at each testing or practice session. " Each treatment must precede and follow each of the other treatments an equal number of times. 18 6
Table 6-3 Within-Group Counterbalancing for the Two-Cola Challenge When Six Participants Are Tested Counterbalancing! Complete counterbalancing! All possible treatment sequences are presented.! You can calculate the number of sequences by using the formula n! (n factorial).! Incomplete counterbalancing! Only a portion of all possible sequences are presented. 20 Table 6-6 An Incomplete Counterbalancing Approach 7
Counterbalancing! Sequence or Order Effects! Sequence or order effects are produced by the participant s being exposed to the sequential presentation of the treatments. " The sequence or order effect depends on where in the sequential presentation of treatments the participant s performance is evaluated, not which treatment is experienced.! Carryover Effects! The effects of one treatment persist or carry over and influence responses to the next treatment. 22 8