Essay Instructions: (Lab 11) Final Project Review & Questions ***This is the final Guideline to my paper
Step 1
First identify a research issue that you find interesting (I hope this isn?t difficult ). It can be any topic that has psychological content or psychological consequences. The specific research you will focus on should consist of only 2 variables that represent either a cause and an effect (e.g., soldiers? wartime trauma on psychological well-being) or an expected relationship (e.g., intelligence and mating behavior). Often the former type of relationship deals with group differences and is explored with experiments (i.e., control everything between groups except for the IV manipulation, then measure the DV between groups), and the latter type of relationship deals with individual differences is explored with correlations. The length of your paper should be 10-12 pages of main text (10 pts off if it is less than 10 full page or greater than 12 pages). Main text refers to the body of the paper (not your references section, not the cover page, and tables/figures aren?t required but if you put them in there they do not count as main text).
? Make sure you have only 2 variables: 1 IV→1 DV:
o Good Example:
The effects of divorce on children?s well-being
o Bad Example:
The effects of divorce and alcoholism on children?s well-being
Why bad? 2 IV?s
The effects of divorce on children?s well-being depending on how old the children are when parents get divorced
Why bad? Again 2 IV?s
The effects of divorce on children?s well-being and future relationships
Why? 2 DV?s
In other words, keep it very simple. Your two-variable study may be influenced by other variables, which you may want to acknowledge (e.g., divorce may affect children?s well-being for boys differently than for girls), but these ?third variables? won?t be measured in your study.
Step 2
Perform a review of the literature using the electronic resources available from the library:
Point your web browser to http://er.lib.msu.edu/. Then under ?All Resources Arranged by Subject? go down to ?Psychology? and click on GO. You can then use databases such as PsychInfo, and you can browse electronic journals. Skim the journal titles for topics that interest you, or some journals that contain a wide range of literature reviews are Psychological Review, Psychological Bulletin, and Psychological Science.
This will help you to identify previous literature that has investigated the issue you are interested in researching (or related issues). If you find the literature has already found an answer to the question, you might still be able to extend what has already been researched. In other words, maybe you see that you are able to propose additional research that will develop a more complete and accurate answer to the research question. Alternatively you could look for an entirely different research area, and see whether that literature has addressed a question in that area you want to propose. The TAs can give you some general guidance if you are completely stumped (they won?t give you a topic, you have to find it).
? Use the library sources from the psych databases
o Best sources:
Journal articles: usual scientific tests of ideas similar to yours that can help you refine your ideas
Recent journal articles: easier to get full-text or in library
Major journals: more likely to be available
o Not as good sources:
Books from psych databases: long reads, not usually scientific studies
Dissertation abstracts: not held in library, difficult to get
o Bad sources (unacceptable):
TV programs, newspaper and magazine articles: subject to more biases, often not scientific studies
Internet information: not subject to validation or review, credibility often questionable
? Ways to improve search:
o Use psychological terms for names of variables of interest
For instance, if you are interested in the effects of how children were raised and the outcome on adult relationships, enter terms like ?attachment style?
o When you find a good article, look at their references to find other good articles
? You will collect 3-5 relevant articles and use the specific information from what you read in the articles, and the general information you have learned in lab, to begin to write your introduction to your topic. Remember that the references must come from established scholarly psychology journals and/or books ? NOT from the Internet, television, radio, magazines, etc.
Here is an example of a reference to a journal article:
Dinkus, B., Petes, G. O., & Kuhn, D. (2000). Memory in toddlers: A review. Journal of Consulting and Clinical Psychology, 112, 56-73.
?and here is an example of a reference to a book:
Ghiselli, E. E., Campbell, J. P., & Zedeck, S. (1981). Measurement theory for the behavioral sciences. San Francisco, CA: Freeman.
Remember that to get full credit in providing your references: (1) they need to be in APA format as shown above, (2) you need to provide copies of all the materials you use: the full article, or if you use a book, please provide stapled copies of the title/author/table of contents and relevant book pages.
By October 31, you should have searched and reviewed the literature. If you want your lab instructor to review your topic, you may do that between October 15 and October 31.
Step 3
Develop your research question into a falsifiable scientific hypothesis consisting of both a null and an alternative hypothesis.
? Best way to state hypotheses:
o After you have described why your research is important, what your ideas are and how they fit in with what has been done before, specifically state the hypotheses being tested at the end of the introduction
o Good example:
H0: There is no relationship between divorce and children?s well-being in the population
H1: There is a negative relationship between divorce and children?s well-being in the population
Bad ways to state hypotheses:
o In the middle of the introduction
o Not specifically: ?I am interested in the relationship between divorce and well-being.?
This is a research idea that can be stated earlier in the introduction, but it is not a hypothesis
o Directionals as hypotheses:
H0: Divorce and well-being will be negatively related
H1: Divorce and well-being will be positively related
The null hypothesis should state that there is no relationship in the population
The alternative hypothesis states that there is some sort of relationship in the population. Usually your theory will suggest the nature/direction of the relationship in your alternative hypothesis. However, if your theory is weak and you?re just exploring the data, then perhaps your alternative hypothesis is not directional, and it simply says there will be statistically significant differences between groups (in either direction) or a statistically significant correlation (either positive or negative)
Only include your 2 variables in your hypotheses; don?t include any potential confounding variables (?third variables? mentioned before) because you are not testing for these (even though you might acknowledge their potential influence in the text)
Step 4
Now that you have identified a research topic, performed a literature review, and formulated scientific hypotheses, it is time to construct your introduction. This section should probably be about 1/3 the total length of your paper.
By November 14, you should have the introduction written. If you want your lab instructor to review the introduction, you may do that between November 1 and November 14.
What to include in your introduction:
Research idea in first paragraph: ?I am interested in studying the effect of psychotherapy on depressed people in urban populations.?
State why your research is important:
o ?Psychotherapy could potentially prevent serious problems (Stevens, 1958).?
Use citations whenever you are stating any facts. Why?
Include information on what past research has found in this area based on your literature review:
o ?Psychotherapy has proven effective to depressive populations in suburban (Smith, 1986) and rural areas (Baker, 1999).?
What your study can add to the research in the area
Hypotheses
What not to include in the introduction:
Sweeping general statements:
o ?Everyone knows that the media influences our perceptions.?
Don?t make claims that can?t be backed up by previous research or are too broad to apply to your study
Opinion and personal statements that add little to your research idea:
o ?This topic is very interesting to me because when I was younger it made a big impact on my life.?
? Your methods and results: save until later
? Don?t mention you research design or what statistical tests you will use until you get to the Methods section
Step 5
Now that you have completed the introduction where you reviewed the existing literature and presented your hypotheses, it is time to develop the methods that you will use to evaluate the research question. In this section of the paper you should describe all aspects of your research design. How will you design your research to address the research question? Will you use an observational design, a true experiment, or a quasi-experimental design? If you use an experimental or quasi-experimental design, how will you manipulate the independent variable to examine its effects on the dependent variable? How many levels of the independent variable will you use?
? Explicitly state your research design:
? Good example:
The effects of anxiety were examined on math test performance. Participants were randomly assigned to receive an anxiety manipulation or a control condition. Anxiety was manipulated by informing participants that the results of the test would determine future income, while those in the control condition were simply told to try their best. A randomized two-group experimental design was used.
? Bad example:
Some participants in this study were made anxious and others weren?t and then took a test. I think that the anxious ones might do worse.
Step 6
Next, you should describe your sampling plan for the experiment in detail. To what population do you hope to generalize the results of your research? Who will be the participants in your research? How will you sample them? What will be the conditions for their participation (extra credit, money, etc). How will you ensure that you do not violate their rights as human subjects in your research?
? Good example:
One hundred fifty undergraduates from a large Midwestern university participated for partial course credit. Participants were recruited by experimenters from large lecture classes. This was a convenience sample. Half of the participants were females while half were male. One hundred were white, forty were African-American, and ten were Asian-American. They ranged in age from 18-27 years with a mean of 20.4 (standard deviation = 2.1 years). These demographic figures closely match those of the university and therefore results will likely generalize to the university population. Participants filled out surveys anonymously and were assured that their responses would be kept confidential and their names would never be reported with any of their data.
? Bad example:
Students filled out surveys in the study and didn?t have to put their names on them. They were all college students in psychology.
Step 7
Now that you have a research design and sampling plan in place for your research, how will you measure the relevant variables for your research? If you use an observational design, you will need to measure both the independent and the dependent variables. If you use an experimental or quasi-experimental design, you will manipulate the independent variable and measure the dependent variable. You should use the literature to identify existing measures of the constructs in your research. If you find measures in the literature, you should report any reliability and validity data available to support the use of the measures. If you cannot find existing measures in the literature, you will need to discuss the process by which you would develop the measures and give an example item or process to demonstrate the measure. You would then need to discuss how you would collect reliability and validity data to support the use of the measures in your research.
? Report all measures that you will use.
? If you are using existing measure:
? List a few sample items ? if you can?t do that (if you can?t find the item content), then give a few examples of what the items might be like.
? Cite it.
? Report reliability statistics such as test-retest or internal consistency.
? Report validity statistics such as correlations to other measures of the same construct (convergent validity)
? If reliability or validity information is not available for an existing measure, describe how you would obtain such information.
? If you are creating your own measure:
? List a few sample items.
? Describe what type of a measure it is (e.g. Likert scale, semantic differential scale)
? State the number of items and how scores will be calculated (e.g. summated rating scale)
? Describe how you would obtain both reliability and validity statistics.
Step 8
Identify the statistical procedure that you will use to evaluate your research hypothesis. Will you use correlation/regression techniques, t-tests, or the Analysis of Variance to statistically analyze your data?
? Refer to the lecture and discuss with your TA: Which statistical tests are used with which design? Which should you use for your design?
Step 9
At this point, you can write up your Method section consisting of the research design, sampling plan, measurement processes, and statistical analysis. This section should require approximately 1/3 the total length of the paper to complete.
By December 2, you should have the methods section written. If you want your lab instructor to review the methods section, you may do that between November 15 and December 2.
Step 10
Now, you won?t actually collect data to test your research hypothesis. However, I?d like you to first assume that your results failed to support your research hypothesis. What might have gone wrong in your research to yield this result?
When you don?t get the results you?ve predicted, there are a few possibilities:
1. The Null Hypothesis is true: There is no relationship between your 2 variables
? This does not require discussion
2. Problems with the research design or manner in which the data was collected may have eliminated the effects of a genuine relationship:
? Sampling could have been ineffective: Describe how
? The manipulation may have been too weak or manipulated something other than the independent variable: Describe how
? The measures may have been invalid or unreliable. This is more likely when using new measures or those measures previously reporting low validity or reliability
? Random assignment could have failed
? The sample size was too small
? The results may only apply in particular settings
? Other ideas??
Step 11
Finally, assume that all went well and your results support your research hypothesis. What are the internal and external threats to validity that might lead you to draw the wrong conclusion about your research results?
When you get the results that you?ve predicted, there are a few possibilities:
1. The Null Hypothesis should be rejected: there is a relationship between the 2 variables in the population.
? You don?t need to discuss this.
2. One or more threats to validity accounts for the relationship between the 2 variables (we have covered threats in earlier labs, so review those materials):
? External validity - Although these results were obtained for the sample, they may not hold for the population. Why?
? Internal validity - One of the many threats may have interfered with the relationship between the independent and the dependent variable. Describe any likely possibilities depending on the design such as regression to the mean, testing, experimenter expectancy effects, history, etc.
Step 12
Complete the assignment by writing up steps 10 and 11 (call Step 10 ?Lack of Support for Research Hypothesis? and Step 11 ?Support for Research Hypothesis). This section should require the remaining 1/3 the length of the entire paper.
