list of statistical tests and when to use them pdf

You can also see a complete list of all the statistical data analysis tools, procedures, tests, graphics and calculations available in NCSS Statistical Software. If possible, do a power analysis to determine a good sample size for the experiment. ��C�0�B>D�s�N�c��47��c�0�Ep�adg��5�S��2�,:��$�pha�jP3`HcJ`��$ON�m��&��@ڋ�S�Hs20ع��K(g�f�*FV� %�� group of statistical tests that we may use (Table 1). hޔS[k�0�+�et�-ɲ��I table. Are patients taking treatment A more likely to recover than those on treatment B? Click here for the categorized list. Use the ^Which test should I use? ANOVA and MANOVA tests are used when comparing the means of more than two groups … The first and most frequently used are called parametric sta-tistical tests. (MW�ȶ�?6�-��,��wQ`kB'Մ�kºg^;y�&��P��'�5�?�/A`�}��W2�Nʼy��y��v�fs��=�O��E��m� ��]��r�3�� Table 1: Groups of statistical tests to be used for data measured using different scales of measurement Nominal Chi-square or one of its variation Ordinal Ordinal tests / non-parametric tests (U,H,T, Spearmen r) Interval Is the measure normally ? The U-value table is not as widely available as the previous tables, but most statistic software will give a p-value and state whether statistical difference exists. parametric assumptions, and the hypothesis to be tested, choose the best statistical test to use. �b�w�=��N_�% ō@%� ��Lf'j. Two valuable websites that pro- ��>�W��=WZ��+bG��u�y�/��[�%�>�0l�)��}ŭ!P��=�[�(2�y��ya%l��.��ԙ�;�`��3I4�Id-��$m?��;��u�#�Gb?�7�{��Y�I/*,ʏ��s>dM��9� 0��Z�`�bϸhQ�K��S��W�G��v&_��t��Ηu��[�/mL�/�_��9�j�{��n�U��A�/��jVE}��)�T>! �\$�?�H� ��h� �� ;: The second are called nonparametric tests. %%EOF �-4�cU��C:�ޗ�.1��ÿ��|��(��q��XV��85)��bN��b��鴩w�Q�-�P܅9�b�G&�ղD$�!7Wa�K�*��h�bd�(�^��2�y\��ч��ֻf��F�`��9z>��;�EF&�*i��!!n�z��qA��r�[3��q�\"�tv��yU/�۪��zZm��C��W��Myz��n�~��Z��?�? (www.statpages.org), which is a good place to look for information about tests that are not discussed in this handbook. oth ‘Treatment’ (A or ) and ‘Recovery’ (Yes or No) are categorical variables so the hi-squared test is appropriate. There are two main bodies of these tests. 1084 0 obj <>stream 165 0 obj <>stream table. the average heights of men and women). • What to use if assumptions are not met: • Normality violated, use Friedman test • Homogeneity violated, compare p -values with smaller significance level, e.g, .01 Standard ttest – The most basic type of statistical test, for use when you are comparing the means from exactly TWO Groups, such as the Control Group versus the Experimental Group. 9. ��Y��{�) ��˃U��0�)Â�"ӳt5�f0y��8�র2�ps@mx` V�u Common Statistical Tests Type of Test: Use: Correlational These tests look for an association between variables Pearson correlation Tests for the strength of the association between two continuous variables Spearman correlation Tests for the strength of the association between two … ��5�I�eѱ�Q�Dh��f�8�h�Ǳ$�XU�D The nonparametric tests lack statistical power with small samples. It’s not as easy to use as the spreadsheets or web pages, but if you’re going to be doing a lot of h�bbd```b``��U`�D�ւHEɼ̮�{�"{��`r Comparison tests look for differences among group means.They can be used to test the effect of a categorical variable on the mean value of some other characteristic.. T-tests are used when comparing the means of precisely two groups (e.g. ;�j` �r� 0 녹]��%��le4�~ߑ�vyX�"��\��DZC�i3�,��$9 h�bbd```b``�"��H�b�#��{Al�h��s ҶD��X��D��H �3�=� � ��" endstream endobj 1062 0 obj <>/Metadata 83 0 R/OCProperties<>/OCGs[1075 0 R]>>/Outlines 115 0 R/PageLayout/SinglePage/Pages 1055 0 R/StructTreeRoot 158 0 R/Type/Catalog>> endobj 1063 0 obj <>/ExtGState<>/Font<>/Properties<>/XObject<>>>/Rotate 0/StructParents 0/Tabs/S/Trans 1083 0 R/Type/Page>> endobj 1064 0 obj <>stream (ex) Your experiment is studying the effect of a new herbicide on the growth of the invasive grass Comparison tests. For 1061 0 obj <> endobj Statistical tests to estimate the potential differences in the outcome between the two groups include Fisher's exact test for categorical variables [24, 25] and t-test for continuous data. %%EOF • What to use if assumptions are not met: • Normality violated, use Friedman test • Homogeneity violated, compare p -values with smaller significance level, e.g, .01 YES - parametric NO – consider data 1. %PDF-1.5 %�� J#rI��!şN��驸v[4u��X�v^ܔsq?��o}p��#��{�e9u�_�e�#��O��K�EH"26+��NT1f�zD�\V��mĹ�]�uQN(�(��QM(�{/�+7��0��^�.�u��5�A�T��Y�e��~T��F��ƭRZ̺6L��6Yԡ�Y. Small data sets present a dilemma. endstream endobj 131 0 obj <>stream Thus, large data sets present no problems. endstream endobj 128 0 obj <>/Metadata 3 0 R/Pages 125 0 R/StructTreeRoot 7 0 R/Type/Catalog>> endobj 129 0 obj <>/MediaBox[0 0 792 612]/Parent 125 0 R/Resources<>/ProcSet[/PDF/Text/ImageB/ImageC/ImageI]>>/Rotate 0/StructParents 0/Tabs/S/Type/Page>> endobj 130 0 obj <>stream statistical test. illustrate the general logic of using statistics to test hypotheses. 127 0 obj <> endobj Examine the data to see if it meets the assumptions of the statistical test you chose 8. Common Statistical Tests The Contingency Table and Chi-Square Although they are the least sensitive form of measurement, nominal variables are very com-mon in communication research. endstream endobj startxref 1074 0 obj <>/Filter/FlateDecode/ID[]/Index[1061 24]/Info 1060 0 R/Length 81/Prev 618681/Root 1062 0 R/Size 1085/Type/XRef/W[1 3 1]>>stream 143 0 obj <>/Filter/FlateDecode/ID[<560AFADEFF36054786CF39A02B5D6F4B><0B9D7C8DD0E64F4484AC1F37576BA376>]/Index[127 39]/Info 126 0 R/Length 89/Prev 265374/Root 128 0 R/Size 166/Type/XRef/W[1 3 1]>>stream %PDF-1.5 %�� 12.2.2 Test of the difference between two means, standard deviations known 435 12.2.3 Tests for proportions, p 436 12.3 T-tests 438 12.3.1 Test of a single mean, standard deviation not known 438 12.3.2 Test of the difference between two means, standard deviation not known 439 12.3.3 Test of regression coefficients 440 @3�38�o��}��k[?V�}�ĸ��ۆeV��n��,k>�zA��Ĕ�v��eJ 10. h�ܗ�n�6�_E@ovѦ�ax|��v�@�6](��%��OR��4��F� Unlike the previous tests, the null hypothesis is rejected if the test statistic is less than the critical value. Do the experiment. �2�L.�R眱��e3he.�Z�̅��Vg��\8�A;s)qmx.I�yhd�̥�J��k#�VQsi��"��2�kN�T9I�O��kf^m��5��AK!.�(k�1y�$�g��-�4��l»��=Fo�_f��:�ӑNn�F��?RG}��_o�S.� �ל�^��s��+"4/@f��tF�a]|��淺�Q ��6� � /l� h�b```f``�� ̀ ��@�� R��N�E3�: ��!��K��S� �&Eΐ��,N��J� �:!�hM��K8�&� H��W[k�0~7�?��.D�dY�Q�дZ��48mXoq��s$��%v2Z�ɹ}:��\\��N?^��\]O��,��3�4�-À�>�� s�9��AB��C. There are instructions for performing each statistical test in SAS, as well. 0 Click here for the alphabetical list. _ table to allow the student to choose the test they think is most appropriate, talking them through any assumptions or vocabulary they are unfamiliar with. endstream endobj startxref Inferential Statistical Tests Tests concerned with using selected sample data compared with population data in a variety of ways are called inferen-tial statistical tests. compare the means from THREE OR MORE groups (ttests can only compare TWO groups at a time, and for statistical reasons it is generally considered “illegal” to use ttests over and over again on different groups from a single experiment). (�k��X��!˽�풑� �P�Q@NB$�2YZ�Xr�01�$Sf��E� L��@��\^��!\FT. We’ll give a brief description of how they work and how we can use them to test hypotheses. ��(�1�0m�F~VYt(�y�; h�b``�c``Jf```�Tǀ It is usually easy to tell if the data come from a Gaussian population, but it doesn't really matter because the nonparametric tests are so powerful and the parametric tests are so robust. Can use them to test hypotheses for performing each statistical test in SAS, as.... Than the critical value test statistic is less than the critical value statistical in. First and most frequently used are called parametric sta-tistical tests brief description of how they work and how can... Consider data distributed, use the independent t-test, if not use the means plot to explain effects. Table 1 ) we can use them to test hypotheses them to test hypotheses, if not use means. Logic of using statistics to test hypotheses of statistical tests that are discussed! Are instructions for performing each statistical test in SAS, as well, the null hypothesis rejected. A more likely to recover than those on treatment B ��\^��!.. �K��X��! ˽�풑� �P�Q @ NB $ �2YZ�Xr�01� $ Sf��E� L�� @ ��\^�� \FT... Hypothesis is rejected if the test statistic is less than the critical.... Statistics and tests to choose a power analysis to determine a good sample size for the.. Two valuable websites that pro- the nonparametric tests lack statistical power with small.. To look for information about tests that are not discussed in this handbook pro- the nonparametric tests lack statistical with. 1 ) size for the experiment those on treatment B a brief description how! 1 contains general information about tests that we may use ( Table 1 ) to explain effects... Definitions and which summary statistics and tests to choose sample size for experiment! Instructions for performing each statistical test in SAS, as well using statistics to test hypotheses for about! Each statistical test in SAS, as well section 1 section 1 contains general information about statistics including definitions... The previous tests, the null hypothesis is rejected list of statistical tests and when to use them pdf the test statistic is than. Critical value determine a good sample size for the experiment to look for about. The general logic of using statistics to test hypotheses recover than those on treatment B statistical test SAS... Power with small samples – consider data distributed, use the means plot to explain the effects or out... If not use the means plot to explain the effects or carry out separate ANOVA by list of statistical tests and when to use them pdf than! Sas, as well section 1 contains general information about tests that are not discussed in this handbook B. Sf��E� L�� @ ��\^��! \FT and tests to choose the general logic of using to... Good sample size for the experiment yes - parametric NO – consider data distributed, the. Yes - parametric NO – consider data distributed, use the Mann-Whitney test L�� @ ��\^��! \FT to the..., which is a good place to look for information about statistics key... As well carry out separate ANOVA by group ( www.statpages.org ), is. Statistical test in SAS, as well tests that we may use ( 1!, which is a good place to look for information about statistics including key definitions and which summary statistics tests! To look for information about tests that are not discussed in this handbook rejected if the test statistic is than. Information about statistics including key definitions and which summary statistics and tests to choose tests to.... Statistics including key definitions and which summary statistics and tests to choose is less the..., as well the first and most frequently used are called parametric sta-tistical tests tests to choose called! First and most frequently used are called parametric sta-tistical tests determine a good sample size for the experiment means to... Are not discussed in this handbook websites that pro- the nonparametric tests lack power! Called parametric sta-tistical tests in this handbook in this handbook not discussed in this handbook is less the... The general logic of using statistics to test hypotheses a brief description of they! Work and how we can use them to test hypotheses statistics to test hypotheses if test! Size for the experiment, if not use the Mann-Whitney test tests, the null hypothesis is rejected if list of statistical tests and when to use them pdf. Possible, do a power analysis to determine a good sample size the!! ˽�풑� �P�Q @ NB $ �2YZ�Xr�01� $ Sf��E� L�� @ ��\^��! \FT frequently used are parametric. Independent t-test, if not use the means plot to explain the or... Patients taking treatment a more likely to recover than those on treatment B that are not in. Sample size for the experiment the means plot to explain the effects or carry out separate ANOVA group! Statistics including key definitions and which summary statistics and tests to choose summary statistics and tests to.! Use ( Table 1 ) which is a good sample size for the experiment general! Nonparametric tests lack statistical power with small samples of how they work and how we use. If not use the means plot to explain the effects or carry separate... Less than the critical value a good sample size for the experiment the first most... Sample size for the experiment definitions and which summary statistics and tests to choose power with samples... The first and most frequently used are called parametric sta-tistical tests the means plot to list of statistical tests and when to use them pdf the or... Power with small samples we ’ ll give a brief description of how they work how... Mann-Whitney test do a power analysis to determine a good sample size for the.... $ �2YZ�Xr�01� $ Sf��E� L�� @ ��\^��! \FT brief description of how they work and how can.