Experiments are research projects where a manipulated independent variable is followed by measuring a change in a dependent variable. There must be at least two groups in experiments that were randomly assigned and treated exactly alike. If a researcher takes into account association, direction of influence, and eliminates rival explanations, then there are only two explanations: the relationship is causal in nature or, the results occurred by chance. Tests of statistical significance are used in order to measure the likelihood that a result of an experiment could have occurred due to chance. Statistical significance is also known as the t-statistic, or the probability that differences are due to chance is less than .05. Otherwise stated, researchers generally accept values of .05 which means that the probability of a relationship occurring due to chance is only 5 times out of 100.
Slide 2: Variations in Experiments, Pt. I
While experiments must always include comparison groups, random assignment, and dependent variable assessment, experiments can be structured in different ways. The three variations in experiments discussed here will be posttest only control group designs, pretest-posttest control group designs, and factorial designs.
Posttest only control group designs are the most basic experimental designs. In these experiments, the dependent variable is measured after the experimental manipulation is applied. In pretest-posttest control group designs, the dependent variable is measured both before and after the experimental manipulation in order to allow for comparisons of the dependent variable. Researchers can apply multiple posttests at differing times to see how time affects dependent variable manipulation, or can measure changes in the dependent variable after making various manipulations. Lastly, the third manipulation of experimental design, are factorial designs, which is when two or more independent variables are manipulated. Factorial designs allow evidence of impact for each individual factor manipulated, as well as a measurement of the factors joined together.
Slide 3: Variations in Experiments, Pt. II
On top of manipulations that researchers can use in experimental designs, manipulations in the experimental context can also influence results. For example, laboratory experiments are those that occur in a controlled environment, which allow for a more targeted examination of the effects of manipulation, but are less realistic as there are no intervening variables present that may exist in the natural environment. Survey-based experiments are when participants are given different, randomly assigned versions of survey questions, however in this design while measurement is usually valid and reliable, the participant is considering hypothetical scenario’s and questions rather than actually living the experience which may bias their results.
Field experiments are when experiments are conducted in a natural setting, which addresses the sterile environment issue of laboratory experiments, but decreases the ease of examining the effects of manipulation. Audit studies are a particular type of field experiments, where matched pairs of participants both participate in an experiment, but only one of the pair is randomly assigned the manipulation, thus allowing for comparison between the pairs.
Slide 4: Variations in Experiments, Pt. III
Quasi-experimental designs are those that resemble experiments, but do not include random assignment. These designs are often sued when random assignment is not possible or feasible in a given population. Quasi-experiments have more validity than not attempting to apply experimental methods, but have less validity and explanatory power than pure experimental studies.
Examples of quasi-experimental designs include: nonequivalent control group designs which are similar to pretest-posttest experimental designs, however, the groups are not randomly assigned and are instead just considered to be similar to one another, before-and-after designs which is when a group is followed over time with pretests and posttests implemented after an intervention or manipulation is applied in order to measure outcomes, and lastly, ex post facto control groups designs, which is when an independent variable is already present or not present in the participant groups, and a dependent variable is measured, meaning that there is not random assignment or manipulation.
When developing and designing studies, researchers must be cognizant of the various benefits and drawbacks of different methodologies and manipulations in order to make sure their research is unbiased as possible.
Slide 5: Validity
The validity of a research design is also affected by the different manipulations applied during experiments. Thus, when considering a project, you must always consider internal and external validity, and how to reduce the threats to these variables.
Internal validity is evidence that rules out the possibility that factors other than the manipulated independent variable are responsible for the measured outcome. This is why it is important to always rule out the possibility of extraneous variables that need to be considered in an environment before conducting a study. Threats of internal validity include not controlling for extraneous variables, selection bias when participants are not randomly assigned, maturation, or the a natural psychological or physiological change that takes place in participants that may change their reactions, and history, which is when environmental factors other than the experimental manipulation change a participants reaction.
External validity is the extent to which experimental findings may be generalized to other settings, measurements, populations, and time periods. One of the goals of research is that findings can be applied outside of an individual study, so considering how to manipulate study designs to increase external validity is important. External validity can be increased by making sure results that happen in a laboratory also occur in a natural setting, by replicating studies, or using a different sample of participants potentially in different settings to see if the same results occur, and being thorough with sample selection.
While it may be impossible to make internal and external validity perfect, being honest about limitations and doing everything in your power as a researcher to increase internal and external validity will increase the impact of your study.
Slide 6: Module Wrap-Up
After reading the texts and listening to the lecture prepared for this module, you should be confident in your ability in completing the learning objectives from the unit.
You should be able to examine casual explanations for attitudes, behaviors, and events and additionally, explain the difference between causation and correlation. Remember that correlation does not imply causality. You should be able to justify the value of experiments and differentiate between the various manipulations to experiments that may need to be applied when doing research. Lastly, you should be able to explain the importance of internal and external validity while identifying the treats to these validities in experiments and describe the different methods that overcome threats to validity.
Your third assignment will be due at the end of this module, and will consist of creating a literature review for your final project this semester. In this assignment, you should provide some justification for why your topic is important for further study and should also identify and review at least 6 of the major studies that have addressed your topic in the past. Make sure to check blackboard or the syllabus for further guidelines regarding this assignment and do not hesitate to post questions in to the interactive discussion board for feedback from your classmates or professor.
JoseRivera123
script4.docx
0
