What are some examples of manipulating variables

What is an experiment?

The aim of the experiment is to check previously formulated statements through a systematic check for their truthfulness by consciously and actively changing individual conditions of the experimental setup. The active manipulation makes up the essential character of the experiment compared to other methods of data collection. The experiment differs significantly from passively oriented methods such as pure observation.

An experiment always consists of certain things Factors, also includes:
1. the examined object or subject
2. the observer
3. the methods of the investigation or the experimental set-up
4. the process of the investigation itself.

In order to guarantee the informative value of an experiment, some conditions be respected:

  • In psychological experiments, the test person (1.) can be subject to fluctuations with regard to certain psychological factors, e.g. they can be particularly motivated precisely because they are in an experiment situation. The same applies to the observer or test leader (2.), who can influence the result through certain expectations (“Rosenthal effect”).
  • With regard to the examination methods (3.), the Replication The following must be taken into account: The results of an experiment must be verifiable, i.e. the conception of an experiment must be designed in such a way that this experiment can be repeated at any time.
  • The investigation itself (4th) finally has to be analyzed after the experiment, which is usually done by the evaluation of the results and data found happens.

The most important elements of an experiment are undoubtedly those variablesin which a distinction is made between dependent, independent and so-called interfering variables. Under independent variables one understands that element of the experiment which is actively and consciously influenced in order to possibly be able to measure a change. The effects of these changes affect the dependent variableswhose influence is to be checked by an experiment. As Disturbance variables Finally, one denotes elements within the experimental set-up that also influence the dependent variables and thus falsify the result of the.

An example:
If an experiment is to investigate to what extent the time of day influences the ability to concentrate, then the time is the independent variable and the subjects' ability to concentrate is the dependent variable. Interfering variables could be, for example, coffee enjoyment, state of motivation or the subject's current state of fatigue.

In general, one can say: the better the confounding variables can be controlled, the greater the significance of the effect of the independent variable on the dependent variable. For the example this would mean: The more you limit the influence of coffee, tiredness or different motivation, the more likely the time of day is responsible for potential changes in the ability to concentrate.
One then speaks of a high internal validity of the experiment, the informative value of this experimental set-up is therefore very high. Opposite is the external validity, with which one describes the validity of the results outside of the experimental setup. If the external validity is high, the experiment is highly representative.

In order to be meaningful at all, however, some important quality criteria must be met when experimenting, without which no real validity can be achieved. A distinction can be made between main criteria and secondary criteria (see Lienert 1989):

To the Main criteria counting:

  • objectivity : The results of the experiment must come about independently of the experimenter, otherwise the measurement method is not meaningful.
  • Reliability : It indicates how high the reliability of the measurement method is and whether the value found can be recorded again after repeated measurements.
  • validity : This criterion states whether and how precisely what was actually measured was what should also be measured. This is often difficult to determine, e.g. when complex factors such as intelligence or creativity are to be assessed in psychological experiments.

There are also a few Secondary criteriawhich, however, do not necessarily all have to be fulfilled in order to be able to speak of a successful experiment:

  • Economy : An experiment is economically viable if it uses more than it costs.
  • usefulness : Indicates the practical, social or scientific use of the data found
  • Standardization, comparability : The measurements should be standardized, i.e. comparable with the results of other objects or test persons.

Especially in the social sciences there is Limits for this method. For example, the test situation can create such an artificial atmosphere that the data found hardly allow any relevant conclusions to be drawn. In addition, there are clear moral and ethical boundaries: An experiment becomes problematic if the test subject is harmed, misled with regard to the intention of the investigation or, for example, in their abilities. B. is manipulated by the experimental setup. There are also high demands on an experiment with regard to data protection.
The experimenter has an extremely responsible task in the experiment, both in dealing with the people involved in the investigation and in terms of the validity of the data found.

  • Hypertext on the theoretical basis of the psychological experiment with Werner Stangl
  • At the e-teaching.org topic special "E-Learning Research", an online expert panel dealt with experimental methods in e-learning research, among other things.
Last change: 01/27/2016

Print preview

create a PDF

e-teaching.org (2016). What is an experiment? Last changed on 01/27/2016. Leibniz Institute for Knowledge Media: https://www.e-teaching.org/didaktik/qualitaet/experiment/index.html. Accessed on May 23, 2021