STAK  _:    ݌vZW V""""""""̪3̪3 @"""""D"DĀ hC0&0 "U""U"U"UU"Uwwww݀UUUUH0 D9D"AA"A>"t"%2d$LA*@ @ "t"G"q]APPa~ `on openStackglobal oldLevelput the userLevel into oldLevelset the userLevel to 2end openStackon resumeStackglobal oldLevelput the userLevel into oldLevelset the userLevel to 2end resumeStackon suspendStackglobal oldLevelset the userLevel to oldLevelif the freesize of this stack 10000 thendomenu "compact stack"end ifend suspendStackon closestackglobal oldLevelset the userLevel to oldLevelif the freesize of this stack 10000 thendomenu "compact stack"end ifend closestackfunction mixran x-- this function will take a series of symbols as input and-- randomly select one of the symbols and return it as the output.-- the symbols must be a list of items, i.e., each symbol must-- be separated by the next in the list by a comma.put the number of items of x into yput random(y) into zput item z of x into zreturn zend mixranon MAST8_$'):3 W=dp?FREEFree Object @BKGD _ BKGD  _r&"=qTitleL(~;ՠ Indexon mouseUpgo first cardend mouseUpBKGD  _r&"=qTitleL(~;ՠ Indexon mouseUpgo first cardend mouseUplP0  PP00 [File New Stac@CARD d.x.@S AdvertisementHow to OrderbThe complete version of the Stat Stack offers many more valuable statistics. These include: Chi-square test with more than one subject category Cochran Q, Wilcoxon, and Walsh tests Komolgorov-Smirnov goodness of fit tests Wald-Walfowitz runs test Spearman rank-order correlation measure Friedman one-way analysis of variance for ordinal scale data Kruskal-Wallis analysis of variance for ordinal scale data Parametric t-test Random number generator and Z-score conversionTo receive the complete Stat Stack, send a check for $15 to:Michael Kelly4025 Lasher RoadDrexel Hill, PA 19026 USAordering informationP0 , PA 19104-6196 USAordering informationP0 FREEFree Object the complete version, press the "How to Order the Complete Version" button.VERSIONS OF THE "STAT STACK":The "Stat Stack" comes in two forms: A standard Hypercard 2.2 stack that requires Hypercard in order to be run and a "stand-alone" version that can operate without Hypercard. This sample version is a "stand alone" application. However, both versions are provided if the complete Stat Stack is purchased. (The stack version saves considerable disk space.)USING THE "STAT STACK":In order to use a particular statistic, press the desired button on the left. This action will take you to the appropriate card for entering data and running the target program. Brief descriptions of each test, as well as the format in which data should be entered, are presented as well. For more detailed descriptions of the statistics, Siegel (1956) is recommended. For most of the tests, significance values are provided with p generally set at .05 (two-tailed where relevant). Some of the programs allow you to select a desired p from a set of options; others require you to test for significance by consulting a table. Relevant tables can also be found in Siegel (1956). Tables will also need to be consulted if a p cutoff other than that provided is desired.REFERENCE:Siegel, S. (1956). Nonparametric statistics for the behavioral sciences. New York: McGraw-Hill.Index@BMAPV 9MM" ]#$"5X$"5pU$5&f(<5Ld6 E|B?:V`N Dp0LB8`$S$$&dCv|%7o$d߽S>S0s?`$0$"d3"BNSB@ (~c 2|h x yEt{~8C`3 <8> |yp|3` {~;bi RZ@€f+  ~D Bo ,* XT* 85p  *2` =e (* Cp`b 4"2@@2Hb L4CC!؀P@!" Sx` "{r H@"3b@>Cx$xc AB@)PMPRR #w( 4YP22bA ()CD~f&NB< 0`xa  n>|?"S!p5w*"`" 0$b@` C5;U"(5pU!p5X V!p]"(U"(6 a`\!p6d1 !  W320 ' xW ')H0&l""8]@6p`CARD"?d j.o Experimental( Control* ) Combined,3R Categories.n Ranked_scores, Rank_groups& RanksF^ Moseson mouseUpif card field 1 is empty or card field 2 is empty thenanswer "Enter scores for the two groups in fields"exit mouseupend ifrepeat with n=3 to 7put empty into card field nend repeatput the number of items of card field 1 into nEput the number of items of card field 2 into nCrepeat with n=1 to nEput item n of card field 1 into xput x&"," after card field 3put "E," after card field 4end repeatrepeat with n=1 to nCput item n of card field 2 into yput y&"," after card field 3put "C," after card field 4end repeatput the number of chars of card field 3 into n_char_3put the number of chars of card field 4 into n_char_4delete char n_char_3 of card field 3delete char n_char_4 of card field 4repeat until card field 3 is emptyput the number of items of card field 3 into all_totput item 1 of card field 3 into smallput item 1 of card field 4 into group_smallput 1 into zrepeat with n=2 to all_totput item n of card field 3 into compareput item n of card field 4 into comp_groupif comparenum_e thenanswer "number of entries in two tables are not equal"exit mouseupend ifput 0 into chirepeat with m=1 to num_oput item m of card field "observed" into obsput item m of card field "expected" into expeput (obs-expe)^2/expe into iterput chi+iter into chiend repeatput num_o-1 into dfset numberformat to "0.###"if p_value is .05 and df 30 thenput item df of card field "p05" into cutoffif chi cutoff thenput "significant chi-square = "&chi&", df = "&df&", p .05"elseput "non-significant chi-square = "&chi&", df = "&df&", p > .05"end ifelse if p_value is .01 and df 30 thenput item df of card field "p01" into cutoffif chi cutoff thenput "significant chi-square = "&chi&", df = "&df&", p .01"elseput "non-significant chi-square = "&chi&", df = "&df&", p > .01"end ifelseput "chi-square = "&chi&", df = "&df&&"(df beyond table range)"end ifend mouseUp,\9 description)(? clearon mouseUpput empty into card field "observed"put empty into card field "expected"end mouseUp$h p05$h p01Chi-Square 1If a population consisting of k categories is randomly sampled, and each of the observations assigned to one of the k categories, the one sample chi-square statistic determines whether the observed frequencies in the different categories matches some expected frequency distribution. In many cases, the researcher expects equal frequencies across the categories, and the program here will calculate the expected value in this case. For cases in which the expected frequency differs across categories, the user must enter the expected frequencies. The program will provide the chi-square value and the degrees of freedom (df). The user can set the p-value to .05 or .01, and the program will indicate whether the chi-square is or is not significant at the desired p-value. If other p-values are desired, the user must consult a chi-square table. No p-values are provided if the number of categories exceeds 30.3.84,5.99,7.82,9.49,11.07,12.59,14.07,15.51,16.92,18.31,19.68,21.03,22.36,23.68,25.00,26.30,27.59,28.87,30.14,31.41,32.67,33.92,35.17,36.42,37.65,38.88,40.11,41.34,42.56,43.776.64,9.21,11.34,13.28,15.09,16.81,18.48,20.09,21.67,23.21,24.72,26.22,27.69,29.14,30.58,32.00,33.41,34.80,36.19,37.57,38.93,40.29,41.64,42.98,44.31,45.64,46.96,48.28,49.59,50.89chi-square 1BMAP Vp#T0@#T D `yp 0pyD"q"yDˆ  P "J 聈2#    " S`ppq%x60@#6 S `30py5z"y5   uI "J 聈e2#  " Da@qx8 CARDdF\u Runs Test-Aon mouseUpask "enter the number of runs in your sample"put it into runsask "enter size of subsample 1"put it into n1ask "enter size of subsample 2"put it into n2put runs - (((2*n1*n2)/(n1+n2)) + 1) into numeratorput 2*n1*n2 into aput a*(a-n1-n2) into aput ((n1+n2)^2)*(n1+n2-1) into bput sqrt(a/b) into denominatorput numerator/denominator into Zset numberformat to "0.###"if Z 1.96 or Z -1.96 thenput "significant Z = "&Zelseput "non-significant Z = "&Zend ifend mouseUp(T2 Recordiv Runs Test-Bon mouseUpif card field "record" is empty thenanswer "Enter results in far right field"exit mouseupend ifask "Enter symbol used for category 1"put it into sym1ask "Enter symbol used for category 2"put it into sym2put the number of items of card field "record" into mput 0 into runsput 0 into n1put 0 into n2repeat with n = 1 to mif item n of card field "record" = sym1 thenput 1+n1 into n1end ifif item n of card field "record" = sym2 thenput 1+n2 into n2end ifif item n of card field "record" <> item n+1 of card field "record" thenput runs+1 into runsend ifend repeatput runs - (((2*n1*n2)/(n1+n2)) + 1) into numeratorput 2*n1*n2 into aput a*(a-n1-n2) into aput ((n1+n2)^2)*(n1+n2-1) into bput sqrt(a/b) into denominatorput numerator/denominator into Zset numberformat to "0.###"if Z 1.96 or Z -1.96 thenput "runs = "&runs&"; significant Z = "&Zelseput "runs = "&runs&"; non-significant Z = "&Zend ifend mouseUp`8LN Clearon mouseUpput empty into card field "record"end mouseUp, I Description Runs TestSuppose a series of samples are taken from a population that consists of two categories X and Y. If the series is arranged in their order of selection, one will observe both alternations between the two categories, as well as some clumping in which a number of members of the same category appear in a row. The runs test determines whether the series contains more alternation or clumping than would be expected by chance. Significant results would therefor indicate that the samples were not drawn randomly from the population. Runs test-A should be used if the number of X observations, Y observations, and runs in the sample are known. If you would like these values calculated, then enter the sequence of observations in the table to the right, separating each observation by a comma. After the observations are entered, press the Runs Test-B button. The p-value for this test is set at .05, two-tailed.Runs TestFREEFree Object d.D.@S AdvertisementHow to Order.The complete version of the Stat Stack offers many more valuable statistics. These include: Chi-square test with more than one subject category Wilcoxon, Walsh, and Cochran Q tests Komolgorov-Smirnov goodness of fit tests Wilcoxin test Wald-Walfowitz runs test Spearman rank-order correlation measure Kruskal-Wallis analysis of variance for ordinal scale data Parametric t-test Random number generator and Z-score conversionTo receive the complete Stat Stack, send a check for $15 to:Michael Kelly4025 Lasher RoadDrexel Hill, PA 19026ordering information`STBL @FTBLGenevaChicagoCourier@PRNT:  0 VH@@D@D@D@D0@D" @H@D"@@!@D"xAq!@DDBRJH$!)@DDBJH$!/@DD(:doMenu "Print Field...","File"6 ͞PRSTHH@R(,, `,-^&'ACARDpd  (o  Group_A(  Group_B F^ Mann-Whitneyon mouseUpif card field "group_A" is empty or card field "Group_B" is empty thenanswer "Enter scores for the two groups in fields"exit mouseupend ifput the number of items of card field 1 into n1put the number of items of card field 2 into n2repeat with n=1 to n1put item n of card field 1 into xput x&"," after card field 3put "A," after card field 4end repeatrepeat with n=1 to n2put item n of card field 2 into yput y&"," after card field 3put "B," after card field 4end repeatput the number of chars of card field 3 into n_char_3put the number of chars of card field 4 into n_char_4delete char n_char_3 of card field 3delete char n_char_4 of card field 4repeat until card field 3 is emptyput the number of items of card field 3 into all_totput item 1 of card field 3 into smallput item 1 of card field 4 into group_smallput 1 into zrepeat with n=2 to all_totput item n of card field 3 into compareput item n of card field 4 into comp_groupif compareqput 1+denom into denomput numer+extra into numerput extra+1 into extraput item extra of card field 5 into qput n+count into countput loops + 1 into loopsend repeatput numer/denom into rankif loops = 1 thenput rank&"," after card field 7elserepeat with y = n to (n+(loops-1))put rank&"," after card field 7end repeatend ifput n+loops into nend repeatput the number of chars of card field 7 into n_char_7delete char n_char_7 of card field 7put 0 into R1put 0 into R2put the number of items of card field 7 into rank_totrepeat with n=1 to rank_totput item n of card field 6 into group_xput item n of card field 7 into rank_xif group_x is "A" thenput rank_x + R1 into R1elseput rank_x + R2 into R2end ifend repeatput (n1*n2)+((n1*(n1+1))/2)-R1 into U1put (n1*n2)+((n2*(n2+1))/2)-R2 into U2put U1 into Uif U2 p) or (p is "x") thenput "non-significant U = "&U&", p > .05"elseput "significant U = "&U&", p .05"end ifelseput U-((n1*n2)/2) into numerput sqrt(((n1*n2)*(n1+n2+1))/12) into denomput numer/denom into z_scoreif z_score 1.96 thenput "significant U = "&U&", p .05"elseput "non-significant U = "&U&", p > .05"end ifend ifend mouseUp* ! Combined*) H Group_ID(m  Group_A( Group_A*  Rankings| -xCܠ Clearon mouseUprepeat with n = 1 to 7put empty into card field nend repeatend mouseUp( N Cutoffs, M$ DescriptionMann-Whitney Ux,x,x,x,x,x,x,x,x,x,x,x,x,x,x,x,x,x,0,0x,x,x,x,x,x,x,0,1,1,1,2,2,2,3,3,3,4,4,4x,x,x,x,0,1,1,2,3,4,5,5,6,7,7,8,9,9,10,11x,x,x,0,1,3,3,4,6,7,8,9,10,11,12,14,15,16,17,18x,x,x,x,2,3,5,6,9,11,12,13,15,16,18,19,20,22,23,25x,x,x,x,x,5,6,8,12,14,16,17,19,21,23,25,26,28,30,32x,x,x,x,x,x,8,10,15,17,19,21,24,26,28,30,33,35,37,39x,x,x,x,x,x,x,13,18,20,23,26,28,31,33,36,39,41,44,47x,x,x,x,x,x,x,x,21,24,27,30,33,36,39,42,45,48,51,54x,x,x,x,x,x,x,x,x,27,31,34,37,41,44,48,51,55,58,62x,x,x,x,x,x,x,x,x,x,34,38,42,46,50,54,57,61,65,69x,x,x,x,x,x,x,x,x,x,x,42,47,51,55,60,64,68,72,77x,x,x,x,x,x,x,x,x,x,x,x,51,56,61,65,70,75,80,84x,x,x,x,x,x,x,x,x,x,x,x,x,61,66,71,77,82,87,92x,x,x,x,x,x,x,x,x,x,x,x,x,x,72,77,83,88,94,100x,x,x,x,x,x,x,x,x,x,x,x,x,x,x,83,89,95,101,107x,x,x,x,x,x,x,x,x,x,x,x,x,x,x,x,96.102,109,115x,x,x,x,x,x,x,x,x,x,x,x,x,x,x,x,x,109,116,123x,x,x,x,x,x,x,x,x,x,x,x,x,x,x,x,x,x,123,130x,x,x,x,x,x,x,x,x,x,x,x,x,x,x,x,x,x,x,138The Mann-Whitney U Test is the most useful non-parametric analogue of the parametric t-test. Like the t-test, it determines whether two independent groups have been drawn from the same population. However, it does not demand the same set of stringent assumptions that the t-test requires. The data, however, must be on an ordinal scale. The user should enter the scores of two independent groups in the appropriate table, with commas separating each score. The program will then determine whether the central tendencies of the scores differ more than would be expected by chance, with p set at .05, two-tailed. The test basically examines the relative ranks of the scores across the two groups, and determines whether the ranks of one group are systematically lower or higher than the ranks of the other group. Note that the program does not correct for ties in ranks. Ties within a group, however, have no effect on the statistical outcome. Ties between scores of different groups can have such effects, but they are negligible. Furthermore, lack of correction produces a more conservative result. Thus, any statistic that is significant without correction for ties would also be significant with such a correction (though not vice versa).Mann-WhitneyFREEFree Object t&"=qTitleN(~;ՠ Indexon mouseUpgo card ID 3035end mouseUpBMAPpV*|!" ! !0"` !(3? `r*F#pC*Ax(P ((1 )@C |4(`P(HP((ᢅ>῁!W !G@ !0W0`` (3?`B*F#0pC*Ax10a00`0a1 V@C |4( ````?ᢅ> CARDd _ T\x Chi-square 1on mouseUpgo card "chi-square 1"end mouseUpV x Runs Teston mouseUpgo card "runs test"end mouseUpTx Binomialon mouseUpgo card "binomial"end mouseUpNx Moseson mouseUpgo card "Moses"end mouseUp,P.descriptionx3G How to Order the Complete Versionon mouseUpgo card "ordering information"end mouseUp\ x Mann-Whitneyon mouseUpgo card "Mann-Whitney"end mouseUp*Qo5iBASIC DESCRIPTION:The Stat Stack contains a set of non-parametric statistics programs (and the parametric t-test) for use with data that cannot meet the assumptions required for parametric tests like pearson r and ANOVA. Some of the tests are useful for data on nominal scales (e.g., the chi-square tests), whereas others can be used when the data are on an ordinal scale (e.g.,Mann-Whitney U test). This version provides a sample of some of the statistics offered on the complete version. For information about the complete version, press the "How to Order the Complete Version" button.VERSIONS OF THE "STAT STACK":The "Stat Stack" comes in two forms: A standard Hypercard 2.2 stack that requires Hypercard in order to run and a "stand-alone" version that can operate without Hypercard. This sample version requires Hypercard, and will work with any 2.X version of Hypercard. However, both versions are provided if the complete Stat Stack is purchased. (The stack version saves considerable disk space.)USING THE "STAT STACK":In order to use a particular statistic, press the desired button on the left. This action will take you to the appropriate card for entering data and running the target program. Brief descriptions of each test, as well as the format in which data should be entered, are presented as well. For more detailed descriptions of the statistics, Siegel (1956) is recommended. For most of the tests, significance values are provided with p generally set at .05 (two-tailed where relevant). Some of the programs allow you to select a desired p from a set of options; others require you to test for significance by consulting a table. Relevant tables can also be found in Siegel (1956). Tables will also need to be consulted if a p cutoff other than that provided is desired.REFERENCE:Siegel, S. (1956). Nonparametric statistics for the behavioral sciences. New York: McGraw-Hill.Index `FREEFree Object eshow menuBarshow card button "make stimuli"show card button "make file"answer "EXPERIMENT OVER!"end mouseUp( Ea warning2 Make Fileon mouseUplock screengo card "stimuli"if card field "stimuliB" is not empty thenput empty into card field "stimuliB"end ifput the number of items of line 1 of card field "stimuliA" into nItemsif nItems <> 1 thenanswer "Stimulus field is empty or contains more than one word per line."exit mouseUpend ifput card field "stimuliA" into wordListrepeat foreverput empty into card field "stimuliA"answer file "Choose file name or cancel to exit"if the result is "Cancel" then exit repeatput it into fileNameopen file fileNamerepeat foreverread from file fileName for 16834if it is empty then exit repeatput it into tempput temp after card field "stimuliA"end repeatclose file fileNameput the number of lines of wordList into nStimrepeat with i = nStim down to 1put line i of wordList into searchFreqfind word searchFreq in card field "stimuliA"if the result is empty thenput word 2 of the foundLine into flput line fl of card field "stimuliA"&return after card field "stimuliB"delete line i of wordListend ifend repeatend repeatdelete last char of card field "stimuliB"ask file "Enter file name for storage of stimulus list"put it into storeFileopen file storeFilewrite card field "stimuliB" to file storeFileclose file storeFileput empty into card field "stimuliB"put empty into card field "stimuliA"go card "display"unlock screenend mouseUp6^gkt@JConsider the words "rain" and "snow." Which one do you think is used more frequently in English? (Press the button labeled "guess" to find out the correct answer.) Human beings are quite good at answering questions about the relative frequency with which different events occur in their environment. In class, we will discuss the importance of this ability and its role in perception and cognition. For now, you'll get a chance to examine this ability for yourself. On each trial of this experiment, you will be presented with pairs of words, letters, or phrases. Your task is to decide which member of the pair is more common. After the experiment, you'll find out how accurate you were, and how closely related your frequency judgments are to the actual frequency of linguistic events. Before beginning the experiment, you must first get some stimuli to judge. You can obtain stimuli in three ways: (1) Drawing items from pre-existing files, (2) Drawing items from files that you create, and (3) Entering items directly on the stimulus card and allowing the computer to retrieve their frequencies from a file that you specify. Each of these methods will be described in turn.Pre-existing files: At present, five files are available for judging word frequencies: brown1, brown2, nouns.brown, verbs.brown, and adj.brown. The "brown" part of the file names refers to the source of the frequency data, namely a well-known word frequency tabulation conducted by researchers at Brown UnivPAGEdE"?w[^W}={z?{S os6LIST3d TAILNu r det slut