PSY202-221640688
From PsychWiki - A Collaborative Psychology Wiki
Concepts:
Dependent Variable
Definition:The variable that is measured.
Example:http://archives.cnn.com/2000/HEALTH/children/09/05/ear.infections/index.html
Application:In the study, the experimenters tested the extend of ear infections caused by pacifiers. Ear infections were the measured variable, therefore ear infections are the dependent variable.
Nominal Variable
Definition:Mutually exclusive and exhaustive categories differing in some qualitative aspect.
Example:http://www.latimes.com/features/health/la-he-milk19-2009oct19,0,486524,full.story
Application:The study explains the different types of milk, which are the exclusive categories, and it explains the differences of each.
Ordinal Variable
Definition:Has the properties of a nominal scale; but in addition, the observations may be ranked in order of magnitude (with nothing implied about the difference between adjacent steps on the scale).
Example:http://www.military-quotes.com/ranks/air-force-rank-insignia.htm
Application:The example shows the military ranks in the air force, which are in order of magnitude.
Interval Variable
Definition:An interval variable, has all the properties of an ordinal scale, such that equal differences between its values represent equivalent intervals.
Example:http://www.big-boards.com/board/934/
Application: The graph has an interval scale, the second graph shows a measurement scale that has an equal distance of one months thus making it an interval variable.
Ratio Variable
Definition:A ratio variable, has all the properties of an interval variable, and also has a clear definition of 0.
Example:http://www.webmd.com/healthy-aging/news/20091202/running-helps-a-healthy-life
Application:In this study, the ratio variable is time, because the people who ran more had a healthier life, and the one who did not run had more health problems. The study has an absolute ratio because some ran nothing which is zero.
Frequency Distribution (regular, grouped, relative, or cummulative)
Definition:Shows the number of observations for the possible categories or score values in a set of data.
Example:http://www.barringtonhigh.org/Document%20Library/SAT%20Report%202008.pdf
Application:The example shows the frequency distribution table of the SAT scores for the selected school.
Percentile (percentile or percentile rank)
Definition:A point on the measurement scale below which a specified percentage of the cases in a distribution falls. Percentile rank: The percentage of cases in a distribution that falls below a given point on the measurement scale.
Example:http://www.homesurfer.com/schoolreports/view/schoolreports.cfm?LEAID=0634620
Application: The school report shows how much it is spent per pupil at the San Juan Unified school, the amount spend is 8564, and their percentile rank in the state is 51.4.
Histogram
Definition:Graph that consists of a series of rectangles, the heights of which represent frequency or relative frequency.
Example:http://daytonarock.com/stats/oct%20stats275.jpg
Application: This shows a histogram that compares the frequency of pages file and hits of the website daytonarock throughout the year month by month.
Frequency Polygon
Definition:Graph that consists of a series of connected dots above the midpoint of each possible class interval (height of the dots corresponds to frequency or relative frequency).
Example:http://www.techcrunch.com/2009/01/22/facebook-now-nearly-twice-the-size-of-myspace-worldwide/
Application:The graph shows the growth of facebook throughout time showing the increase of frequency of users over myspace, therefore making it a frequency poligon.
Bar Diagram
Definition:Used for qualitative data, a graph that is similar to a histogram, except that space appears between the rectangles.
Example:http://www.dmp.gov.bd/static/crime_watch.php?year1=2009%20&%20month1=08
Application:The bar diagram shows the types of crime commited on the x axis, and the frequency on the y axis.
Pie Chart
Definition:Used for qualitative data, area in any piece of the pie shows the relative frequency of a category.
Eexample:http://www.cdfa.ca.gov/Statistics/
Application:The pie chart shows distribution and relative frequency of the California agricultural production.
Mean
Definition:Sum of all the scores divided by the total number of scores.
Example:http://www.barringtonhigh.org/Document%20Library/SAT%20Report%202008.pdf
Application:The report shows the mean for each of the subject SAT scores.
Median
Definition:Value that divides the distribution into halves; another name for P50.
Example:http://www.barringtonhigh.org/Document%20Library/SAT%20Report%202008.pdf
Application: Taken the critical thinking scores on the report the median score would be 550 to 599.
Mode
Definition:Score that appears with the greatest frequency.
Example:http://www.barringtonhigh.org/Document%20Library/SAT%20Report%202008.pdf
Application:The score that had the most recordings which were a total of 39 for students was 500-599, therefore that score is the mode.
Range
Definition:Difference between the lowest score and the highest score.
Example:http://www.barringtonhigh.org/Document%20Library/SAT%20Report%202008.pdf
Application:The SAT scores vary from scores of 200 to 800, therefore the range is 600.
Variance
Definition:Mean of the squares of the deviation scores
Application:In this article they show the different variance for each type of baseball player. The average variance is .300.
Standard Deviation
Definition:Square root of the variance.
Example:http://www.istockanalyst.com/article/viewarticle/articleid/3693752
Application:In the artucle they talk about the standard deviation of the price of gold. The article says that for the past 10 years, one standard deviation for gold is plus or minus 7.3 percent.
Standard Scores (z-scores)
Definition:States how far away a score in from the mean in standard deviation units: one type of standard score.
Example:http://joongangdaily.joins.com/article/view.asp?aid=2913642
Application:In this article, they show how a student is now evaluated by the standard scores relative to the others who took the exam. It shows what countries have a higher standard score than other countries. College applicants are evaluated according to these standard scores rather than the raw data scores.
Scatterplot
Definition:A graph of a bivariate distribution consisting of dots at the point of intersection of paired scores.
Example:http://www.johnmyleswhite.com/notebook/2008/11/
Application:The scatterplot shows the correlation between suicide rates and GDP by country. The scatterplot has no meaningful correlation between the two variables.
Correlation (r)
Definition:A measure of the degree of relationship between two variables.
Example:http://www.visualizingeconomics.com/2008/04/17/does-higher-income-increases-happiness/
Application: In this article, the two variables being measured are happiness(average life satisfaction) and money. According to the results there is a positive correlation between happiness and money. The more money one makes the happier one is.
EXTRA CREDIT: Correlation does not equal causation
Example:http://www.nytimes.com/2009/12/08/health/research/08real.html
Application:In the US, high cholesterol has increased with the obesity rate, the study says there is a correlation between high cholesterol and hair loss, but the increase of hair loss has also increased because of the use more hair products, and the weather changes, therefore there are many extraneous variables that can affect this correlation.
Guideline 7:Check that results are fairly represented in graphics or concluding statements.
Example:http://www.youtube.com/watch?v=YCbYTrYD5y8
Application:The comparing of both of the cell phone coverage areas in the graphics are misrepresented, to fit the benefit of one of the company.
Guideline 2:Consider the source
Example:http://www.youtube.com/watch?v=YCbYTrYD5y8
Application:The comparing of both cell phone companies coverage favors one because the information is biased to one obviously giving invalidate information.
Guideline 2:Consider the source
Example:http://www.youtube.com/watch?v=igdyXceBZLA
Application:The result of the comparing between Verizon and AT&T can be biased because of the source where it comes from to benefit one side.
Guideline 7:Check that results are fairly represented in graphics or concluding statements.
Example:http://kara.allthingsd.com/files/2009/08/stockcat.jpg
Application:The picture of the cat is suppossed to represent a drecrease in data, but without points the graph misrepresents the data.
Guideline 7:Check that results are fairly represented in graphics or concluding statements.
Example:http://kara.allthingsd.com/files/2009/08/128731636518771160.png
Application:The graph misrepresents data, it does not specify for what population the data is.
Guideline 7:Check that results are fairly represented in graphics or concluding statements.
Example:https://www.mahalo.com/how-to-handle-finances-in-a-depression
Application:The frequency polygon shown cannot be true because it has no x axis or y axis labeling.
Guideline 7:Check that results are fairly represented in graphics or concluding statements.
Example:http://www.iforcenutrition.com/dymethazine/productinfo.html
Application:The graph in this article misrepresents data because it has no y axis values. Also, the graph has pictographs and 3D bars which can be misleading.
Guideline 7:Check that results are fairly represented in graphics or concluding statements.
Example:http://mmmlabs.com/b/wp-content/uploads/2009/04/hd-graph.jpg
Application:the graph has pictographs which can mislead and distract. The pictures are just decorative which are not needed. The simpler the graph the better.
Guideline 7:Check that results are fairly represented in graphics or concluding statements.
Example:http://competition-gigantic.com/
Application:The histogram shows no values for the y axis, which can be misleading to the data.It says after the adding of the data those were the finalist bands, but yet they show no data but only the bars with no numbers.
Guideline 7:Check that results are fairly represented in graphics or concluding statements.
Example:http://monkeyartawards.typepad.com/monkeyartawards/images/2008/05/18/copywriter_graph.png
Application:The proportions of the pencils are not the right size.The size of the pencils are not correspondent to the percentage that they represent, therefore it misleads the data represented.
EXTRA CREDIT: “Lie on Purpose”
Statistics can be misleading in four ways,one of the ways it can be misleading on purpose is by lying on purpose from the researcher part, this involves not collecting data but creating fake data to achieve the results the researcher wants. A person who was caught lying on purpose was Hwang Woo-suk who is a South Korean veterinarian researcher. He was a professor at Seoul National University, but was fired on March 20, 2006. This researcher became badly known for fabricating and faking data within a series of experiments, which were published in high profile journals, in the field of stem cell research. This researcher was best known for his two articles that were published in the journal Science in 2004 and 2005 where he fraudulently reported to have succeeded in creating human embryonic stem cells by cloning. The researcher first caught the world’s attention when he announced he successfully created a cloned dairy cow, but he failed to provide scientifically verifiable data for the research, giving only media sessions. Later in 2004, he claimed to create an embryonic stem cell with somatic cell nuclear transfer method. He later claimed several more cloning successes, including a dog. The suspicion began when one of researchers who had worked with Hwang told the press he was ceasing from the collaboration of the research due to the source of donations of the eggs. The eggs had come from women who were being paid $1400 US. On December 2005, the university found out that all findings oh Hwang’s stem cell were fabricated and faked. The punishment for the researcher for faking data was removing his papers from the journals. He was also sentenced to a two year suspended prison on October 2009, after being found guilty for embezzlement and bioethical violations, but he was cleared of fraud. The only way this could have been avoided was to have found proof from the first time he was avoiding giving and presenting the data. Another way is by before publishing making sure the data is testable and replicable.
◄ Back to Descriptive Statistics PSY 202 Fall 2009 page
