VIEWS: 7 PAGES: 35 POSTED ON: 1/13/2014 Public Domain
Internal Assessment Overview Mr. Freeman 1 Workshop Overview Null and Experimental Hypothesis Design of study Sampling of study Choosing the right Inferential Test for your study Questions/Comments 2 Null Hypothesis The logic of traditional hypothesis testing requires that we set up two competing statements or hypotheses referred to as the null hypothesis and the alternative(experimental) hypothesis. These hypotheses are mutually exclusive and exhaustive. Ho: The finding occurred by chance H1: The finding did not occur by chance 3 Null Hypothesis For testing, you will be analyzing and comparing your results against the null hypothesis, so your research must be designed with this in mind. It is vitally important that the research you design produces results that will be analyzable using statistical tests. 4 Hypothesis testing Most social scientist are very afraid of statistics, due to obscure mathematical symbols, and worry about not understanding the processes or messing up the experiments. There really is no need to fear! Most psychologists understand only the basic principles of statistics, and once you have these, modern computing technology (SPSS, Microsoft excel) gives a whole battery of software for hypothesis testing. Designing your research only needs a basic understanding of the best practices for selecting samples, isolating testable variables and randomizing groups, and choosing the right inferential testing measures. 5 Hypothesis testing A common statistical method is to compare a population to the mean. For example, you might have come up with a measurable hypothesis that IB students have a higher IQ if they take IB Psychology for two years. Your alternative hypothesis, H1 would be “Students who take IB Psychology for two years (IV) will show a higher IQ increase (DV) than students who have not.” Therefore, your null hypothesis, H0 would be “Students who take IB Psychology for two years do not show a higher IQ increase than students who do not.” 6 Hypothesis testing In other words, with the experiment design, you will be measuring whether the IQ increase of student who take IB Psychology deviate from the mean, assumed to be the normal condition. “H0 = No increase. The students will show no increase in mean intelligence.” 7 Hypothesis testing From IQ testing of the control group, you find that the students who did not take IB Psychology (control group) have a mean IQ of 100 before the experiment and 100 afterwards, or no increase. This is the mean against which the sample group will be tested. The students who took IB Psychology show an increase from 100 to 106. This appears to be an increase, but here is where the statistics enters the hypothesis testing process. You need to test whether the increase is significant, or if experimental error and standard deviation could account for the difference. 8 Hypothesis testing Using an appropriate test (an inferential test-depending on the design chosen), the researcher compares the two means, taking into account the increase, the number of data samples and the relative randomization of the groups. A result showing that the researcher can have confidence in the results allows rejection of the null hypothesis. Remember, not rejecting the null is not the same as accepting it. It is only that this particular experiment showed that IB Psychology had no affect upon IQ. This principle lies at the very heart of hypothesis testing. Just because you failed to prove that IB Psych increases your IQ does not mean that you prove that it does not! In other words, just because your experiment failed to proved that IB Psychology increases IQ, does not mean that it is not the case. 9 Significance The exact type of statistical test used depends upon many things, including the field, the type of data and sample size, amongst other things. The vast majority of scientific research is ultimately tested by statistical methods, all giving a degree of confidence in the results. 10 Significance For psychology, the researcher looks for a significance level of 0.05, signifying that there is only a 5% probability that the observed results and trends occurred by chance. The significance level determines whether the null or alternative is rejected, a crucial part of hypothesis testing. 11 Writing a Hypothesis The entire experiment and research revolves around the research hypothesis (H1) and the null hypothesis (H0), so this is a very intricate part of the experiment. Needless to say, it can all be a little intimidating, and many international students have found this to be the most difficult stage of the IA. 12 Example (borrowed from my college professor) A worker on a fish-farm notices that his trout seem to have more fish lice in the summer, when the water levels are low, and wants to find out why. His research leads him to believe that the amount of oxygen is the reason – fish that are oxygen stressed tend to be more susceptible to disease and parasites. 13 Example (borrowed from my college professor) He proposes a general hypothesis. “Water levels affect the amount of lice suffered by rainbow trout.” This is a good general hypothesis, but it gives no guide to how to design the research or experiment. The hypothesis must be refined to give a little direction. 14 Example (borrowed from my college professor-Dr. Jackson-Lowman) Rainbow trout suffer more lice when water levels are low.” Now there is some directionality, but the hypothesis is not really testable, so the final stage is to design an experiment around which research can be designed, a testable hypothesis. Rainbow trout suffer more lice in low water conditions because there is less oxygen in the water.” This is a testable hypothesis – he has established variables, and by measuring the amount of oxygen in the water, eliminating other controlled variables, such as temperature, he can see if there is a correlation against the number of lice on the fish. This is an example of how a gradual focusing of research helps to define how to write a hypothesis. 15 Example (borrowed from my college professor-Dr. Jackson-Lowman) Once you have your hypothesis, the next stage is to design the experiment, allowing a statistical analysis of data, and allowing you to test your hypothesis. The statistical analysis will allow you to reject either the null or the alternative hypothesis. If the alternative is rejected, it is OK! Remember the study is not intended to be ground breaking. This is part of the scientific process, striving for greater accuracy and developing ever more refined hypotheses. 16 Design of your study Repeated measure design Independent measure design 17 Repeated measure design A repeated measures design consists of testing the same individuals on two or more conditions. The key advantage of the repeated measures design is that individual differences between participants are removed as a potential confounding variable. Also the repeated measures0 design requires fewer participants, since data for all conditions derive from the same group of participants. 18 Repeated measure design The design also has its disadvantages. The range of potential uses is smaller than for the independent groups design. For example, it is not always possible to test the same participants twice. There is also a potential disadvantage resulting from order effects, although these order effects can be minimized. Order effects occur when people behave differently because of the order in which the conditions are performed. For example, the participant’s performance may be enhanced because of a practice effect, or performance may be reduced because of a boredom or fatigue effect. 19 Repeated measure design Order effects act as a confounding variable but can be reduced by using counterbalancing. If there are two conditions in an experiment the first participant can do the first condition first and the second condition second. The second participant can do the second condition first and the first condition second and so on. Therefore any order effects should be randomized. 20 Independent measure design If two groups in an experiment consist of different individuals then this is an independent measures design. The main advantage of an independent measures design is that there is no problem with order effects. However, the design also has disadvantages. The most serious is the potential for error resulting from individual differences between the groups of participants taking part in the different conditions. 21 Sampling One of the most important issues about any type of method is how representative of the population the results are. The population is the group of people from whom the sample is drawn. For example if the sample of participants is taken from sophomore IB students, the findings of the study can only be applied to that group of people and not all Rickards High School students and certainly not all people in the world. Obviously it is not usually possible to test everyone in the target population so therefore you will use sampling techniques to choose people who are representative (typical) of the population as a whole. 22 Opportunity Sampling Opportunity sampling is the sampling technique most used by IB psychology students. It consists of taking the sample from people who are available at the time the study is carried out and fit the criteria your are looking for. This may simply consist of choosing the first 20 students in your class to fill in your sample quota. It is a popular sampling technique as it is easy in terms of time. It can also be seen as adequate when investigating processes which are thought to work in similar ways for most individuals such as memory processes 23 Opportunity Sampling However, there are many weaknesses of opportunity sampling. Opportunity sampling can produce a biased sample as it is easy for the researcher to choose people from their own social and cultural group. This sample would therefore not be representative of your target population as you friends may have different qualities to people in general. A further problem with opportunity sampling is that participants may decline to take part and your sampling technique may turn into a self selected sample. 24 Self-Selected Sampling Self selected sampling (or volunteer sampling) consists of participants becoming part of a study because they volunteer when asked or in response to an advert. This sampling technique is used in a number of the core studies, for example Milgram (1963). This technique, like opportunity sampling, is useful as it is quick and relatively easy to do. It can also reach a wide variety of participants. However, the type of participants who volunteer may not be representative of the target population for a number of reasons. For example, they be more obedient, more motivated to take part in studies and so on. 25 Choosing the right Inferential test Choosing the right statistical test may at times, be a very challenging task for a young psychology student. In order to choose the right statistical test, when analyzing the data from an experiment, we must have at least: § a decent understanding of some basic statistical terms and concepts; § some knowledge about few aspects related to the data you collected during the research/experiment (e.g. what types of data we have - nominal, ordinal, interval or ratio, how the data are organized, how many study groups (usually experimental and control at least) you have, are the groups paired or unpaired, and are the sample(s) extracted from a normally distributed. 26 Choosing the right Inferential test your IAs will fall in the following design Most of type: Repeated measures 1 sample or 2 sample Parametric As your advisor, I can suggest inferential test associated with your sample but it is your responsibility to choose the right inferential test. 27 Choosing the right Inferential test also depends on the type of data that Your test you have. The following terms are used are used to describe types of data and by some to dictate the appropriate statistical test to use: Nominal Ordinal Interval Ratio 28 Choosing the right Inferential test Interval Data: Temperature, Dates (data that has an arbitrary zero) Ratio Data: Height, Weight, Age, Length (data that has an absolute zero) Nominal Data: Male, Female, Race, Political Party (categorical data that cannot be ranked) Ordinal Data: Degree of Satisfaction at Restaurant (data that can be ranked) 29 Choosing the right Inferential test Potential inferential test: T-Test Wilcoxon matched pair test Chi Square Mann-Whitney U test. 30 Wilcoxon signed rank sum test The Wilcoxon signed-rank test is a non-parametric statistical hypothesis test used when comparing two related samples or repeated measurements on a single sample to assess whether their population mean ranks differ (i.e. it's a paired difference test). How to use the Wilcoxon signed rank sum test: http://faculty.vassar.edu/lowry/ch12a.html 31 T-Test Use Student's t-test when you have one nominal variable and one measurement variable, and you want to compare the mean values of the measurement variable. The nominal variable must have only two values, such as "male" and "female" or "treated" and "untreated.“ It is usually used to: (1) To test hypothesis about the population mean (2) To test whether the means of two independent samples are different. (3) To test whether the means of two dependent samples are different. (4) To construct a confidence interval for the population mean. 32 Chi square The Chi Square (X2) test is undoubtedly the most important and most used member of the nonparametric family of statistical tests. Chi Square is employed to test the difference between an actual sample and another hypothetical or previously established distribution such as that which may be expected due to chance or probability. Chi Square can also be used to test differences between two or more actual samples 33 Mann-Whitney U test The Mann-Whitney U-test is used to test whether two independent samples of observations are drawn from the same or identical distributions. An advantage with this test is that the two samples under consideration may not necessarily have the same number of observations. 34 Mann-Whitney U test For more info on statistical test: http://www.statsoft.com/textbook/nonparametric- statistics/ http://www.fon.hum.uva.nl/Service/Statistics/Signe d_Rank_Test.html http://statistics-help-for- students.com/How_do_I_report_paired_samples_T_ test_data_in_APA_style.htm 35