Download
1 / 65
650 likes | 723 Vues
The importance of an effect? No, significance depends on sample size. That the null hypothesis is false? No, it is always false. That the null hypothesis is true? No, it is never true. An effect size is a standardized measure of the size of an effect:
E N D
The importance of an effect? • No, significance depends on sample size. • That the null hypothesis is false? • No, it is always false. • That the null hypothesis is true? • No, it is never true.
An effect size is a standardized measure of the size of an effect: • Standardized = comparable across studies • Not (as) reliant on the sample size • Allows people to objectively evaluate the size of observed effect. Andy Field
r = .1, d = .2 (small effect): • the effect explains 1% of the total variance. • r = .3, d = .5 (medium effect): • the effect accounts for 9% of the total variance. • r = .5, d = .8 (large effect): • the effect accounts for 25% of the variance. • Beware of these ‘canned’ effect sizes though: • The size of effect should be placed within the research context. Andy Field
There are several effect size measures that can be used: • Cohen’s d • Pearson’s r • Glass’ Δ • Hedges’ g • Odds Ratio/Risk rates • Pearson’s r is a good intuitive measure • Oh, apart from when group sizes are different … Andy Field
Organization of Reports • There are seven sections to a report. Their headings appear centred on the page. Under some of the major sections, there are subsections which are located at the left margin and are underlined. The seven sections are: Title, Abstract, Introduction (no heading), Method, Results, Discussion, and References, if any. • Title Page • After the short title and page number, the running head should be given. This is left justified in all uppercase letters. Do not exceed 50 characters, including punctuation and spaces. An example: Running head: ATTENTIVE INTERFACES: DESIGNING FOR CREATIVE PROCESS • The title should be a concise statement of the main topic of the report, usually consisting of about 12 to 15 words. It should refer to the major variables or theoretical issues under investigation. Since the purpose of the title is to inform the reader, it should be explanatory when standing alone. Avoid words that serve no useful purpose and only increase the length. Such phrases as "A Study of..." or "An Experimental Investigation of..." should be avoided. Do not use abbreviations in the title. All words should be spelled out for clarity. Centred directly under the title should appear your names and under that your institution (ITU!).
Summarises the whole study, including results and major findings • One page maximum, one paragraph with strict control of allowable number of words in publications etc. • The word "Abstract" is centred on the page. Then a one-paragraph summary of your research report is given. This paragraph is not indented. It should be written last. This paragraph should concisely describe the problem under investigation, the participants, the experimental method, findings, and conclusions. To conserve characters in the abstract, type all numbers except those that begin a sentence as digits.
The introduction does not require a heading; however, the title of the paper should be typed, centred at the top of the first page of the introduction. • A good introduction addresses two questions: What has been done in this area by other researchers? and, What is the point of the present study? • The introduction is the place to include the review of the research literature that led to your hypothesis. For instance, you might show how prior findings are inconsistent or ambiguous. Explain how your experiment may clarify the problem. State your hypothesis explicitly toward the end of the introduction, after you have explained the research and thinking that led to it. Identify independent and dependent variables here. You may want to include a sentence or two about operational definitions (or you can do it in Method). • If you have made predictions about the outcome of the study, say so. Be sure you say why you expect these results. Do not expect readers to guess what you are thinking. In the introduction, you are moving from the general to the specific: a general discussion of the problem area, to your specific hypothesis.
Method - This section must be very detailed and clear. It tells the reader that someone else can repeat the experiment just by reading your method section. The method section generally consists of three subsections: participants, apparatus (or materials), and procedure. A fourth, optional, subsection is design. • Participants - The age, sex, and any other relevant demographic data are presented here. State how many subjects participated, how they were selected, and how they were assigned to groups. • Apparatus or Materials - A description of the apparatus used is given here, including any software used to run the experiment. If materials (such as a questionnaire) were used, either cite your source (if published materials were used) or provide a copy in the appendix of your paper if you devised the instrument yourself. You should describe the instrument in your materials section. For example, a 50-item six-point Likert-type questionnaire was devised by the experimenter to measure attitudes toward authority figures. Half of the questions were worded such that....The highest (positive) score that could be attained on the measure was 300; the lowest (negative) score was 50. Thus, higher scores reflected more positive attitudes toward authority figures. • Procedure - This section describes what the experimenter did and how it was done. It is a detailed description of the events that the experimenter went through from the beginning until the end of the study. Such things as experimental and control group assignment to conditions, order or manner of experimental treatment presentation, and a summary of the instructions to the participants are presented here. Include a statement about your research design and the operational definitions of your variables.
This section is where you present your data and analyses. The experimenter gives a description and not an explanation of the findings of the experiment. In order to fulfill this requirement, the results section should include descriptive statistics (rather than the raw data) and statistical tests if used. Include degrees of freedom used, obtained values of inferential statistics performed, probability level, and direction of effect. Underline letters used as statistical symbols, such as "N", "F", "t", "SD", and "p." (Use underlining, not quotation marks. Since many Web browsers using underlining to indicate a link, avoid underlining within web pages.) Make reference to any figures and tables used, for example, "(see Table 1)."The reference to the table or figure should be close to the relevant material in the text. Never use a figure or table without referring to it in the text. • Tables are often used when presenting descriptive statistics such as means, standard deviations and correlations. Pictures, graphs, and drawings are referred to as figures. You should use as few tables and figures as possible. They should be used as supplements, not to do the entire job of communication. (See the APA manual for detailed guidelines for Tables and for Figures.)
Generally, report descriptive statistics, then inferential statistics, then state in words what was found. • Include graphs, tables of descriptives etc. before results of analysis, one hypothesis after the other. • Include calculations of effect size, degrees of freedom etc.
State your conclusions on the basis of your analyses. The conclusions should be related to the questions raised in your introduction section. How is this study, and these results, relevant to the field? Open the discussion section with a statement of support or non-support for your original hypothesis, in plain words, without the maths. You may want to point out differences or similarities between other points of view and your own. You may remark on certain shortcomings of the study, but avoid dwelling on flaws for too long, if it’s a good scientific study, most of the shortcomings would be examined and eliminated prior to publication! In general, this section allows you relatively free reign to examine, interpret, and qualify your results. Include suggestions for further research: does your study suggest that things may be more complicated than you thought? How might the next researcher clarify that?
Must be listed in a specific format dictated by the journal/publication. For your reports, use the APA format Article From an Online Periodical Author, A. A., & Author, B. B. (Date of publication). Title of article. Title of Online Periodical, volume number(issue number if available). Retrieved from http://www.someaddress.com/full/url/ Bernstein, M. (2002). 10 tips on writing the living Web. A List Apart: For People Who Make Websites, 149. Retrieved from http://www.alistapart.com/articles/writeliving Online Scholarly Journal Article: Citing DOIs Because online materials can potentially change URLs, APA recommends providing a Digital Object Identifier (DOI), when it is available, as opposed to the URL. DOIs are an attempt to provide stable, long-lasting links for online articles. They are unique to their documents and consist of a long alphanumeric code. Many-but not all-publishers will provide an article's DOI on the first page of the document. Article From an Online Periodical with DOI Assigned Author, A. A., & Author, B. B. (Date of publication). Title of article. Title of Journal, volume number, page range. doi:0000000/000000000000 Brownlie, D. (2007). Toward effective poster presentations: An annotated bibliography. European Journal of Marketing, 41(11/12), 1245-1283. doi:10.1108/03090560710821161 Article From an Online Periodical with no DOI Assigned Author, A. A., & Author, B. B. (Date of publication). Title of article. Title of Journal, volume number. Retrieved from http://www.journalhomepage.com/full/url/ Kenneth, I. A. (2000). A Buddhist response to the nature of human rights. Journal of Buddhist Ethics, 8. Retrieved from http://www.cac.psu.edu/jbe/twocont.html Article From a Database For articles that are easily located, do not provide database information. If the article is difficult to locate, then you can provide database information. Only use retrieval dates if the source could change, such as Wikis. Manual.Smyth, A. M., Parker, A. L., & Pease, D. L. (2002). A study of enjoyment of peas. Journal of Abnormal Eating, 8(3), 120-125.
For your reports, appendices should include the survey questions if a survey was used, or the test material if an experiment was conducted/ screenshots of web-content. • List as appendix 1, 2, etc. with a new page for each new item
1. Conduct a literature review and write your introduction. This section explains the problem under investigation, presents the research question(s) and hypotheses and summarizes previous, relevant research. Make sure that every previous research cited in the text with (Surname, X., 2011) there is a full reference in a reference list. Write your introduction, moving from a general description of the research area to the specific questions addressed by your study. Give any previous support for your choice of method, generally, also in the introduction as you get more specific. 2. Compose the methods section of the report. This describes the techniques, processes or tests you used to confirm or reject your hypotheses and why these methods were chosen. The method section includes several subsections to describe the sampling methods, materials, procedure etc. 3. Report your findings. This section contains raw data usually in tables, lists and figures, descriptive statistics, the mathematical expression of the hypotheses and the dependent and independent (predictor and outcome) variables. The analysis chosen for each hypothesis are stated and the assumptions are demonstrated with reference to the descriptives, followed by the test statistic and the mathematical and verbal expression of the results. 4. Discuss or interpret your results relative to the hypotheses. The section may also include a discussion of the significance of the findings and point to future research topics. If there are reasons to doubt the generalisability of your results to the population, place those in the discussion section. 5. Construct the references used in your report. Carefully follow the citation guidelines in the approved style guide (APA). 6. Write the abstract. In publications, the abstract is a key report section because it is often what is provided to information browsers. It should be one continuous paragraph of no more than one page, stating the general research question, one or two sentences about the current thinking or important previous findings relevant to the research question, the hypotheses and results, with mathematical values, and the major important findings of the study. 7. Format headers, subheadings, tables and figures according to the required style guide (APA). 8. Review your draft for typographical errors, orderly presentation, correct and complete references and grammar, and continuity of terminology.
Graphs should (Tufte, 2001): • Show the data. • Induce the reader to think about the data being presented (rather than some other aspect of the graph). • Avoid distorting the data. • Present many numbers with minimum ink. • Make large data sets (assuming you have one) coherent. • Encourage the reader to compare different pieces of data. • Reveal data.
Histograms plot: • The score (x-axis) • The frequency (y-axis) • Histograms help us to identify: • The shape of the distribution • Skew • Kurtosis • Spread or variation in scores • Unusual scores
Boxplots are made up of a box and two whiskers. • The box shows: • The median • The upper and lower quartile • The limits within which the middle 50% of scores lie. • The whiskers show • The range of scores • The limits within which the top and bottom 25% of scores lie
What Does The Boxplot Show? Top 25% Upper quartile Median Middle 50% Lower quartile Bottom 25%
The bar (usually) shows the mean score • The error bar sticks out from the bar like a whisker. • The error bar displays the precision of the mean in one of three ways: • The confidence interval (usually 95%) • The standard deviation • The standard error of the mean
Is there such a thing as a ‘chick flick’? • Participants: • 20 men • 20 women • Half of each sample saw one of two films: • A ‘chick flick’ (Bridget Jones’ Diary), • Control (Memento). • Outcome measure • Physiological arousal as an indicator of how much they enjoyed the film.
How to cure hiccups? • Participants: • 15 hiccup sufferers • Each tries 4 interventions (in random order): • Baseline • Tongue-pulling manoeuvres, • Massage of the carotid artery, • Digital rectal massage • Outcome measure • The number of hiccups in the minute after each procedure
Is text-messaging bad for your grammar? • Participants: • 50 children • Children split into two groups: • Text-messaging allowed • Text-messaging forbidden • Each child measures at two points in time: • Baseline • 6 Months later • Outcome measure • Percentage score on a grammar test
Anxiety and Exam Performance • Participants: • 103 students • Measures • Time spent revising (hours) • Exam performance (%) • Exam Anxiety (the EAQ, score out of 100) • Gender
