This action might not be possible to undo. Are you sure you want to continue?
\
TSTICS FOR INTRODUGTORY
COURSES
J
STATISTICS  A set of tools for collecting, oreanizing, presenting, and analyzing numerical facts or observations. I . Descriptive Statistics  procedures used to organize and present data in a convenient, useable. and communicable form.
f
MEAN The ooint in a distribution of measurements about which the summeddeviationsare equal to zero. Average value of a sample or population. POPULATION MEAN SAMPLE MEAN
2. Inferential Statistics  procedures employed to arrive at broader generalizations or inferences from sample data to populations. l STATISTIC  A number describing a sample characteristic. Results from the manipulation of sample data according to certain specified procedures. DATA  Characteristics or numbers that a r e c o l l e c t e db y o b s e r v a t i o n . J POPULATION  A complete set of actual or potential observations. J J  A number describing a PARAMETER population characteristic; typically, inferred f r o m s a m p l es t a t i s t i c .
p: +!,*,
o:#2*,
Note: The mean ls very sensltlveto extrememeasurementsthat are not balancedon both sides. I WEIGHTED MEAN  Sum of a setof observations multiplied by their respectiveweights, divided by the sum of the weights: 9, *, *, LWEIGHTED MEAN ; : n u m b e ro f w h e r ex r , : w e i g h t , ' x , o b s e r v a t i o nG fd rom a population. grdups.'Calculate observaiion sample.or gr6upings in a frequencydistribution. Ex. In the FrequencVDistribution below, the meun is 80.3: culculatbd by using frequencies for the wis. When grouped, use clossmidpointsJbr xis. in a observation or potenlial J MEDIAN  Observation set that divides the set so that the same number of observationslie on each side of it. For an odd number of values.it is the middle value; for an even number it is the averageof the middle two. Ex. In the Frequency Distribution table below, the median is 79.5. f MODE  Observationthat occurs with the greatest tiequency. Ex. In the Frequency Distributioln nble below. the mode is 88.
tiom O SUM OF SOUARES fSSr Der iations andsummed: the mean. squared (I r,),  li. r x ) ' o r I x i ', t PopulationSS:I(X N _ r, \,)2 SS:I(xi x)2or Ixi2Sample of squaredifferO VARIANCE  The average andtheir mean. observations between ences SAMPLEVARIANCE POPULANONVARIANCE
,\r*'
VARIANCESFOH GBOUPEDDATA POPUIATION SAMPLE
t=1
o2:*i
lI
^{G'{G
;_r
t , ( r ,  p) t s 2 = ; 1 i t i l m '  x ; 2
 A subset of the population f SAMPLE selectedaccording to some scheme. SAMPLE  A subset selected J RANDOM in such a way that each member of the population has an equal opportunity to be selected. Ex.lottery numbers in afair lottery J VARIABLE  A phenomenon that may take on different values.
root of D STANDARD DEVIATION  Square
the variance: Ex. Pop. S.D. o
n Y
I
D BAR GRAPH  A form of graph that uses bars to indicate the frequency of occurrence of observations. o Histogram  a form of bar graph used rr ith interval or ratioscaled variables.  Interval Scale a quantitative scale that permits the use of arithmetic operations. The zero point in the scale is arbitrary.
U
fi
Shows the number of times each observation occurs when the values ofa variable are arranged in order according to their magnitudes.
GROUpITG OF DATA
z
)
J il {il,
I a rrI.)'A .l b], K I 3artl LQ
x 99 98 gl 96 95 94
f
x
t
11 11111 1
x
f
100 1 83
1
74 11f
75 1111
x 65 66 67 68 69
t
o
1 11 1 111
BISTRI. FREOUENCY tr CUMULATUE
thetowhich shows BUTION A distribution tal frequencythrough the upper real limit of eachclass. tr CUMUIATIVE PERCENTAGE DISTRI. BUTIONA distributionwhich showsthetothrough the upper real limit of tal percentage eachclass.
ut
0 85 0 86
11 87
76
77 7A 79
11 111
I
o
1 1111111 111 11 1
0 88 0 89
I
11 1 11 I
80
81
70 1111 71 0
72 11
 R.atio Scale same as interval scale excepl that there is a true zero point. D FREOUENCY CURVE  A form of graph representing a frequency distribution in the form of a continuous line that traces a histogram. o Cumulative Frequency Curve  a continuous line that traces a histogram where bars in all the lower classes are stacked up in the adjacent higher class. It cannot have a negative slop€. o Normal curve  bellshaped curve. o Skewed curve  departs from symmetry and tailsoff at one end.
93
92
0 91
82
73 111
EilSTRIBUTION FREOUENCY II GROTJPED
 A frequency distribution in which the values ofthe variable have been grouped into classes.
I il {.ll
CLASS
6567
lNl.l'tlz
!I!
Cum f
3 11 16 25 31 35
llrfGl:
f I
"
4.84 17.74 25.81
6&70
7173
CLASS 98100
f
CLASS
t
7+76 Tt79 8082
8385 8688 8991 92g 9597
9&100
3 8 5 9 6 4 8 8 6 1 2 2
15 10 0 att?
N O R M A LC U R V E ^/T\ ./ \
t
40.32 50.00
56.45 69.35
\
43
51 57 58 60 62
82.26
91.94 93.55 96.77
15 10
CURVE SKEWED
/ /
J
LEFT
\ \ \ \
100.00
0
RANDOM VARIABLES
Probability of occurrence^t at Number of outcomafamring Ant=@ oif'ent'l EwntA
A mapping 'onlv or function that assignsone and onenumerical value to each outcome in an exPeriment.
of an D SAMPLE SPACE  All possibleoutcomes experiment. N TYPE OF EVENTS
o Exhaustive  two or more events are said to be exhaustive if all possible outcomes are considered. Symbolically, P (A or B or...) l. two or more events are said to be nonrNonExhausdve exhaustive if they do not exhaust all possible outcomes. Exclusive  Events that cannot occur rMutually simultaneously:p (A and B) = 0; and p (A or B) = p (A) + p (B).
Aprobabilin tr LEVEL OF SIGNIFICANCE ution. distrib rare in thesampling value considered oneis where thenull hypothesis under specified of chance theoperation willing to acknowledge
factors. Common significance levels are 170, 5 0 , l 0 o . A l p h a ( a ) l e v e l : t h e l o w e s tl e v e for which the null hypothesis can be rejected. critical region. The significanceleveldeterminesthe [ NULL HYPOTHESIS (flr)  A statement that specifies hypothesized value(s) for one or more of the population parameter. lBx. Hs= a That isp : 0.5.] coin is unbiased. HYPOTHESIS (.r/1)  A tr ALTERNATM statement that specifies that the population parameter is some value other than the one specified underthe null trypothesis.[Ex. I1r: a coin is biased That isp * 0.5.1 HYPOTHESIS I. NONDIRECTIONAL an alternative hypothesis (H1) that states onll that the population parameter is different from the one ipicified under H 6. Ex. [1 f lt + !t0 TwoTailed Probability Value is employed when the alternative hypothesis is nondirectional.  an HYPOTHESIS 2. DIRECTIONAL alternative hypothesis that statesthe direction rn which the population parameter differs fiom the Hf lr ' t1 one specified under 11* Ex. Ilt: Ir > pn rtr OneTailedProbability Value is employedu'hen the alternative hypothesis is directional. PROOF  Stnct D NOTION OF INDIRECT interpretation ofhypothesis testing reveals that thc' null hypothesis can never be proved. [Ex. Ifwe toi. a coin 200 times and tails comes up 100 times. it i s no guarantee that heads will come up exactly hali the time in the long run; small discrepancies migfrt exist. A bias can exist even at a small magnitude. We can make the assertion however that NO BASIS EXISTS FOR REJECTING THE THE COIN IS HYPOTHESIS THAT UNBIASED . (The null hypothesisis not reieued. When employing the 0.05 level of significa reject the null hypothesis when a given res occurs by chance 5% of the time or less.] ] TWO TYPES OF ERRORS
Ex. males, females Exclusive  Events that can occur oNonMutually simultaneously: p (A orB) = P(A) +p(B)  p(A and B)' &x. males, brown eyes.  Events whose probability is unaffected Slndependent by occurrence or nonoccurrence of each other: p(A lB) = p(A); ptB In)= p(e); and p(A and B) = p(A) p(B). Ex. gender and eye color  Events whose probability changes SDependent deoendlns upon the occurrence or nonoccurrence ofeach other: p{.I I bl dilfers lrom AA): p(B lA) differs from p ( B ) ; a n dp ( A a n dB ) : p ( A ) p ( B l A ) : p ( B ) A A I B ) Ex. rsce and eye colon
tl DISCRETE RANDOM VARIABLES  Involvesrulesor probabilitymodelsfor assign(not fraconly distinctvalues ing or generating tionalmeasurements). C BINOMIAL DISTRIBUTION  A model trials for the sum of a seriesof n independent wheretrial resultsin a 0 (failure) or I (success).Ex. Coin to p ( r ) = ( ! ) n ' l  t r l "  ' "t in n wherep(s) is the probabilityof s success trials with a constantn probability per trials, ,1\= n! a"n  dw "h ' e ' r e( t s / s,! ( n  s ) !
Binomial mean: !: nx Binomial variance: o': n, (l  tr) A s n i n c r e a s e s ,t h e B i n o m i a l a p p r o a c h e st h e Normal distribution. DISTRIBUTION D HYPERGEOMETRIC A model for the sum of a series of n trials where each trial results in a 0 or I and is drawn from a small population with N elements split between and N2 failures. Then the probabilN1 successes ity of splitting the n trials between xl successes and x2 failures is: Nl! {_z! p(xlandtrr:W 't 4tlvr;lr + Hypergeometric mean: pt :E(xi
C JOINT PROBABILITIES  Probabilitythat2 ot more eventsoccur simultaneously. tr MARGINAL PROBABILITIES or Uncondi= summationof probabilities' tional Probabilities
 Probability PROBABILITIES D CONDITIONAL of I given the existence of ,S, written, p (Al$. fl EXAMPLE Given the numbers I to 9 as observations in a sample space: .Events mutually exclusive and exhaustive' n nurnbers) Example: p (all odd numb ers) ; p ( all eue .Evenls mutualty exclusive but not exhaustiveExample: p (an eien number); p (the numbers 7 and 5) .Events ni:ither mutually exclusive or exhaustiveExample: p (an even number or a 2)
variance: and o2: ffit+][p]
D POISSON DISTRIBUTION  A model for the number of occurrences of an event x : 0 , 1 , 2 , . . . ,w h e n t h e p r o b a b i l i t y o f o c c u r r e n c e is small, but the number of opportunities for t h e o c c u r r e n c ei s l a r g e , f o r x : 0 , 1 , 2 , 3 . . . .a n d )v > 0 . otherwise P(x) =. 0.
e$t=ff
P o i s s o nm e a n a n d r a r i a n c e : , t .
 Type 1 Error (Typea Error) = the rejectionof 11,whenit is actuallytrue.The probabilityof a type 1 error is givenby a. TypeII Error(Type BError)=The acceptance offl, whenit is actuallyfalse.Theprobabilin of a type II error is given by B.
p re ssed q u en t' i es u re e.t Fo r c ontinuo u s t'a ri u b I es. .fi'e in terms o.f areus under u t'ttt.re.
fl SAMPLING DISTRIBUTION  A theoretical probability distribution of a statistic that would iesult from drawing all possible samples of a
given size from some population.
D CONTINUOUS RANDOM VARIABLES  Variable that may take on any value along an uninterrupted interval of a numberline.  bell cun'e; D NORMAL DISTRIBUTION a distribution whose values cluster symmetrically around the mean (also median and mode). f(x)=1, o"t'2x (xP)212o2
THE STAIUDARDEBROR OF THE MEAN
A theoretical standard deviation of sample mean of a given sample si4e, drawn from some speciJied population. DWhen based on a very large, known population, the error is: standard 6 __ _ o "r ^ ln EWhen estimated from a sample drawn from very large population, the standard error is:
O= ^= tS 'fn
wheref (x): frequency.at.a givenrzalue : s t a n d a r dd e v i a t l o no f t h e o distribution
lt p x
: approximately I 111q 2.7183 approximately : the meanof the distribution : any scorein the distribution
(for sample mean X) rlf x 1, X2, X3,... xn , is a simple random sample of n elements from a large (infinite) population, with mean mu(p) and standard deviation o, then the distribution of T takes on the bell shaped distribution of a normal random variable as n increases andthe distribution ofthe 7! ratio: 6l^J n approaches the standard normal distribution as n goes to'infinity. In practice.a normal approximation is acceptable for samples of 30 or larger. Percentage Cumulative Distribution
for selected Z values under a normal curye
lThe dispersion of sample means decreasesas sample size is increased.
D STANDARD NORMAL DISTRIBUTION  A normalrandomvariable Z. thathasa mean of l. deviation of0. andstandard deviaQ ZVALUES  The numberof standard liesfrom themean: observation tionsa specific ': x 11
Zvalue PercentifeScore
3 2 l 0 +1 +2 +3 o13 2.2a 15.87 50.00 a4.13 97.72 99.a7
Critical when
region for rejection u : OO7. twotailed
of Hs test
.2.b8
O
+2.58
esof a reliable NBIASEDNESS Property
imator beins estimated. o Unbiased Estimate of a Parameter  an estimate that equals on the averagethe value ofthe parameter. Ex. the sample mesn is sn unbissed estimator of the population mesn. . Biased Estimate of a Parameter  an estimate that does not equal on the averagethe value ofthe parameter. Ex. the sample variance calculated with n is a bissed estimator of the population variance, however, x'hen calculated with nI it is unbiused. J STANDARD ERROR  The standard deviation of the estimator is called the standard error. Er. The standard error forT's is. o: = "/F X This has to be distinguished from the STAND.A,RDDEVIATION OF THE SAMPLE:
' The standard error measuresthe variability in the Ts around their expectedvalue E(X) while the stanJard deviation of the sample reflects the variability rn the sample around the sample'smean (x).
tr USED WHEN THE STANDARD DEVIATION IS KNOWN: When o is known it is possible to describethe form of the distribution of must The sample meanasa Z statistic. the sample be drawn from a normal distribution or have a samplesize(n) of at least30. mean(either whereu : population =r ! ',. 6= under Ho) and or = o/f,. nro#rf or hypothesized o Critical Region  the portion of the areaunder thosevaluesof a statistic the curvewhich includes that leadto the rejectionofthe null hypothesis.  The most often used significance levels are using z0.01,0.05,and0.L Fora onetailedtest to zvaluesof 2.33, statistic,thesecorrespond test, For a twotailed 1.65,and 1.28respectively. the critical regionof 0.01 is split into two equal markedby zvalues of 12.581. outerareas Example 1. Given a population with lt:250 and o: S0,whatis theprobabili6t of drawing a sample of n:100 values whosemean (x) is at 2=1.00.Looking atThble least255?In this case, A, the given areafor 2:1.00 is 0.3413. Tb its or 15.85%. right is 0.1587(=6.50.i413) Conclusion: there are spproximately 16 chancesin 100 of obtaining a sample mean : 255 from this papulation when n = 104. Example 2. Assume we do not know the population me&n. However, we suspect that it may have been selectedfrom a population with 1t= 250 and 6= 50,but we are not sure. The hypothesis to be tested is whether the sample mean was selectedfrom this populatian.Assume we obtainedfrom a sample (n) of 100, a sample ,neen of 263. Is it reasonable to &ssante that this sample was drawn from the suspectedpopulation? = 250 (that the actualmeanof the popu . H o'.1t lationfrom which the sampleis drawn is equal to 250) Hi [t not equal to 250 (the alternative hypothesiS is that it is greaterthan or less than 250, thus a twotailed test). the popula2. estatisticwill be usedbecause tion o is known. 3. Assumethe significancelevel (cr)to be 0.01. Looking at Table A, we find that the area beyond a z of 2.58 is approximately 0.005. t}reabTo reject H6atthe 0.01levelof significance, solutevalue of the obtainedz must be equalto or greater thanlz6.91l or 2.58.Herethevalueof z corto sample mean: 263is 2.60. responding tr CONCLUSIONSince thisobtained z fallswithin we mayreject H oatthe0.01level thecriticalregion, of significance.
Table A
.00 o.o ol o.2 o.3 o.4 .ol .o2
Normal Areas Curve
area tron mean to z .o4 .o5 ,06 .o7 .6 .o9
.(x!
o o o o . 0 0 4 0 . 0 0 8 0 . 0 1 2 0 0 1 6 0 0 1 9 9 . 0 2 3 9. 0 2 7 9 . 0 3 1 9 .0359 0 3 9 8 0 4 3 8 . O 4 7 4. O 5 1 7 0 5 5 7 . 0 5 9 6 . 0 6 3 6 . 0 6 7 5 . 0 7 1 4 . 0 7 5 3 0 7 9 3 0 8 3 2 . 0 8 7 1 0 9 1 0 0948 .0987 .1026 .'tO64 .1'lO3 .1141 1 1 7 9 1 2 1 7 1 2 5 5. 1 2 9 3 1554 1591 1624.1664
2.1 2.2 2.3 2.4 2.5
.4452 .4554 .4641 .47't3 .4772 .4821 .4A61 .4893 .4918 .4938 .lgsg .4965 ,gYOC .4974 .4974
.4463 .4474 .4564 .4573 .4649 .46S .4719 .4726 .4774 .4743 .4826.4830 .4A64 .4A68 .4a96 .4898 .4920 .4922 .4940 .4941 .+955 .+ss6 .4966 . +YOO .4967 . {Vot .4975 .4976 .4975 .4976
.4444 .4542 .4664 .4732 .4744 .4834 .4871 .4901 .4925 .4943 assz .4WO .4968 .4977 .4977
.4495 .4591 .4671 .4734 .4793 .+aga .4875 .4904 .4927 .4945 a959 .9VqY .4969 .4977 .4977
.4505 .4599 .4674 .4744 .4798 tetz .4A78 .4906 .4929 .4946 .+goo .+VrV .4970 .4978 .4978
.4515 .460a .4646 .4750 .4803 aaa6 .4aal .4909 .4931 .4944 .+s6r .+rt .4971  .4979 .4979
.4525 .4616 .4693 .4756 .4aAa .taso .4884 .4911 .4932 .4949 .+s6z ,1Vt 4 .4972 .4979 .4979
,4535 .4625 .4699 .4761 .4812 ZSsa .4887 .4913 .4934 .4951 .csi6g .{VtJ .4973 .4980 .4980
.4545 .4633 .4706 .4767 .4417 ABs? .4A90 .4916 .4936 .4952 .4964.+Vr+ .4974 l 1 .4S81 l .4S81
.4981 .49A2 .4982 .49a3 .4984 .4984 .4985 .4985 .4986 .4986 l .4987 .4991 .4987 .4988 .4988 .4989 .4949 .4989 .4990 .4990
DEvIAI r L'sBDwHEN THE STANDARD r. I rtoN IS UNKNOWNUseof Student's is estimated from itsvalue o is notknown, f When
s a m o l ed a t a .
F
f'
jm tratio the ratio employedil thq.testingof a significancebf Ivpotheses or determiningthe case) Vrri'erence betweenmeafrsltwosample inrolving a samplewith a idistribuiion.The tbrmula Ts:
\ F
SX
Example. Given x:l08, s:l5, and n26 estimatea 95% confidence interval for the population mean. Since the population variance is unknown, the tdistribution is used.The resultinsinterval.usins a tvalve of 2.060 fromTable B (row 25 of the middlecolumn), is approximately 102 to 114. Consequently, any hyp between 102 to 114 is tenableon the pothesized basis of this sample.Any hypothesizedprbelow 102 or above 114 would be rejectedat 0.05 significance. O COMPARISON BETWEEN I AND z DISTRIBUTIONS Althoueh both distributions are svmmetrical about a meanbf zero, the fdistribution is more spread out than the normal di stributi on (zdistributioh).
where p : population mean under H6
lrl r =.s "n6 oDistributionsymmetrical distribution with a mean of zero lnd standard deviation that annroachesone as degreesoffreedom increases the Z distribution). 'i.i . approaches and condition required in .A,ssumption r\suming rdistribution: Samplesare diawn from a normallv distributed population and o rpopulation standard deviatiori) is unknown. o Homogeneity of Variance If.2 samples are b e r n c c o m o a r e d t h e a s s u m p t i o ni n u s i n g t  r a t i o r' th?t the variances of the populatioi's from * h e r e t h e s a m p l e sa r e d r a w n a r e e q u a l . o E s t i m a t e d6 X  ,  X ,( t h a t i s s x ,  F r ) i s b a s e do n thc unbiasedestimaie of the pofulaiion variance. o Degrees of Freedom (dJ\^the number of values that are free to vary after placing certain restrictions on the data.
Example. The sample (43,74,42,65) has n = 4. The sum is 224 and mean : 56. Using these 4 numbers and determining deviationsfrom the mean, we'll have J deviations namely (13,18,14,9) which sum up to :ero. Deviations from the mean is one restriction we have imposed and the natural consequence is that the sum ofthese deviations should equal zero. For this to happen, we can choose any number but our freedom to choose is limited to only 3 numbers because one is restricted by the requirement that the sum ofthe deviations should equal zero. We use the equality:
Thus a much larger value of t is required to mark off the bounds of the critical region <if rejection. As d/rincreases, differences between z andt distributions are reduced. Table A (z) may be used instead of Table B (r; when n>30. To Lse either table when n<30,the sample must be drawn from a normal population.
Ta^bJs
'!4
.1 B*=l.evel
1
.
tr CONFIDENCE INTERVAL Interval within which we may consider a hypothesis tenable. Common confidence intervals are 90oh,95oh, and 99oh. Confidence Limits: limits definins the confidence interval. (1 cr)100% confidence interval for rr:
3 4 6
o.o25 o.o5 12.706 4.3Q3 3 . 18 2 2.776 2.571 2.447 2.306 _ ?.?42 22?E 2.201 2..1 2 1 77 99
.Z
d
0.o1 olo2 3r.tz1 6.965 4.541 1747 3.305 3.143 _
2.994
0.oo5 o.ol 6i.657 9925 5.441 4.604 4O3Z 3.707 _
3.499
:e10 . !t ,13 .14 .1 _ _] q 17 18 t 20
.
2.896 2.82L _ 2.76L _ 2.714 2..6 2 6 AA 11
2.650 .
3.355 3.250 3.169 3.106 3..0 3 0 55 55
g.ot z
:0 (x, x) + (x29 + ft t x) + (xax) q given mean of 56, iJ'the 3 observqtions ure So first I 43, 74,und 42, the last observationhus to be 65. This single restriction in this case helps us determine df, Theformula is n lessnumber of restrictions. In this t'ase,it is nl= 4l=3df . _/Ratiois a robust test This meansthat statistical arelikely valid despitefairly largedepartures inferences from normality in the population distribution. If normality of populationdistributionis in doubt,it is wise the samplesize. to increase
z. r oo
ii, *F t lil. l,<i +z * ("1{n)
where Z  , i s t h e v a l u eo f t h e standard normal variable Zthat puts crl2 percent in each tail of the distribution. The confidence interval is the complement of the critical regions. A tstatistic may be used in place of the zstatistic when o is unknown and s must be used as an estimate. (But note the caution in that section.)

,aa
.25 ,?s
24
27
.29  30 ini,
2.
2.145 _ 2.924 2.'131 , 2.60? 2:12o _ ?,qe3 2,110 2.5Q7 2.1o1 2.552 2.Q93_2.5392.OqQ 2.524 2.514 2.OQO 2.5Q8 2Bf4 2.o6s 2.5Oq ?.e64 _2.4e2 2.060 _ 2.1qF 2.479 ?.o50 2.052 2.423 2.044 2.467 2.Q45 ?,482 2.042 2.457
2. _ _ 
?.9f7 2.947 2.921 ?.e98 2.474 461 2.445 2.431 2.a1e 2AA7 ?.1e7 2.7Q7 2.779 2.771 2.763 Z.lsE 2.750
GORRELATION
OF THE N SAMPLING DISTRIBUTION DIFFERENCE BETWEEN MEANS If a number of pairs of samples were taken from the same population or from two different populations, then: r The distribution of differences between pairs of samplemeanstendsto be normal (zdistrjbution). r The mean of these differences between means F1, 1"is equal to the difference between the population means. that is ltfltz. I ZDISTRIBUTION: or and ozure known o The standard error ofthe difference between means o",  ={toi)  \ + @',) I n2 ", o Where (u,  u,) reDresents the hvpothesized differencein rirdan!.'ihefollowins statisticcan be used for hypothesis tests: _ (4tr)(ut uz)
D STANDARD ERROR OF THE DIFFERENCE betweenMeansfor CorrelatedGroups. The seneralformula is:
""r* " 7r zrsrrsr,
where r is Pearsoncorrelation o By matching samples on a variable correlated with the criterion variable, the magnitude of the standard error ofthe difference can be reduced. o The higher the correlation, the greater the reduction in the standard error ofthe difference.
^2
^2
n
Definirton  Carrelation refers to the relatianship baween two variables,The Correlstion CoefJicientis a measurethst exFrcsses the extent ta which two vsriables we related
Q *PEARSON r'METHOD (ProductMoment CorrelationCoefficient)  Cordlationcoefficient variables. employed with intervalor ratioscaled Il Ex.: Givenobservations to twovariablesXand we cancompute their corresponding u values: :(xR)/s" :{yD/tv. Z, andZ, 'The formulas for the Pearson correlation(r): : {* ;Xy y)
JSSI SSJ  Use the above formula for large samples.  Use this formula (also knovsn asthe MeanDevistion Makod of camputngthe Pearsonr) for small samples. 2( z2,,) r:__iL is quicker and can Q RAW SCORE METHOD be used in place of the first formula above when the sample values are available. (IxXI/)
z=
o When n1 and n2 qre >30, substifuesy and s2 for ol and 02. respecnvely.
"r.r,
(j;.#)
(Toqbtain sum of squaygs(SS) see Measures of Central Tendencyon'page l) D POOLED 'TEST o Distribution is normal on<30 r o1 and 02 are zal known but assumed equal  The hvoothesistest mav be 2 tailed (: vs. *) or I is 1rl > lt2 @r 1t, 2 talled:.1i.is 1t, andrhe.alternative p 2 and the alternatrve s y f p2.)  degreesof freedom(dfl : (n 2. rI)+(n y 1)=n fn 2 formula below for estimating 6riz giyen ;U.r11!9 to determrnest,x,.  Determine the critical region for reiection by aslevelof sisnificdnce and[ooksieningan acceptable in! at ihe rtablb with df, nt + i22. o_ Use the following formula for the estimated standard error: (n1 l)sf +(n2 l)s nI+n2 n1*n22 n tn2
(
z
D PURPOSE Indicatespossibility of overall meaneffectof the experimental treatments before investigatinga specific hypothesis. D ANOVA Consistsof obtaining independent It allowsfor from populationsubgroups. estimates the partition of the sum of squaresinto known components of variation. D TYPES OF VARIANCES ' Between4roupVariance(BGV reflects themagnihrdeof thedifference(s) amongthegroupmeans. ' WithinGroup Variance (WGV) reflectsthe group. It is also dispersion within eachtreatment referredto asthe error term. tI CALCULATING VARIANCES ' Following the Fratio, when the BGV is large relativeto theWGV. the Fratiowill alsobe larse. o2(8i'r,,')' 66y=
k1
oMost widelyused nonparametric test. .The y2 mean : its degreesof freedom. oThe X2 variance : twice its degreesof fieedorl o Can be usedto test one or two independentsamples. o The square of a standardnormal variable is a c h i  s q d a r ev a r i a b l e . o Like the tdistibution" it has different distributions depending on the degrees of freedom. D DEGREES OF FREEDOM (d.f.) . / v COMPUTATION o lf chipquare tests for the goodnessof'fitto a hr p o t h e s i z ' ed di s t r i b u t i o n . d.f.: S  I  m, where r in the fiequene . g:.number of groups,or classes. dlstrlbutlon. m  number of population that lnu\r  s a m n l eparameters b e e s t i m a t e df r o m s t a t i s t i c st o t e s t t h c h y p o t h e ss i. o lf chisquara testsfor homogeneityor contingene r d.f : (rows1) (columnsI) f, GOODNESSOFFIT TEST To anolr thtc h i  s q u a r e d i s t r i b u t i o n i n t h i s m a n h ' e r t .h . ' c r i t i c d l c h i  s q u a r ev a l u e i s e x p r e s s e d as: (f_i) ' ,nh".. , : /p observedfreqd6ncy ofthe variable /"  .lp..,tSd fiequency (basedon hypothesrzcd p o p u l a t l o no l s t n D u t r o n ). tr TESTS OF CONTINGENCY Application t'l to two separate populaiionsto te.r Chisquarete.sts s t a t r s t l c ar l n d e o e n d e n co ef a t t r r b u t e s . D T E S T S O F H O M O G E N E I T Y  A p p l i c a t i o no t Chisquaretests to tWpqq.mplqs to_ test'iTthey canrc f r o m f o p u l a t i o n s w i t h l i k e 'd i s t r i b u t i o n s . D R U N S T E S T  T e s t s w h e t h e r a s e q u e n c e( t t r c o m p r i s ea s a m p l e ) . i sr a n d o m .T h e ' f o l l o ui n g equatrons areapplred:
nt'n11 l2ntnr(2ntn, ls . '.t ,rnR ,= r= , ,2 r" ,n ' r1 ' Ir , t "rrJ 1",*"r12 1nr+n,l1 Where
0
(
tU I
: mean of ith treatment group and xr., of all n values across all k treatmeiii
WGV:
.ss,+ss, *...+^ss*
D HETEROGENEITY OF VARIANCES mav by using the Ftest: be determined
o s2lllarger variance'1 stgmaller variance)
D NULL HYPOTHESIS Variancesare equal and their ratio is one. HYPOTHESISVariances TIALTERNATM differ andtheir ratio is not one. f, Look at "Table C" below to determine if the variancesare significantly different from each other. Use degrees of freedom from the 2 nyI). samples:(n11,
"_O (seeMeawherethe SS'sare the sumsof squares sures of Central Tendencyon page 1) of each valuesaroundthe subgroupmean. subgroup's D USING F.RATIO F : BGV/WGV of freedomare kl for the numerator 1 Degrees and nk for the denominator. ' If BGV > WGV, the experimentaltreatments are responsiblefor the large differencesamong group means.Null hypothesis: the group means are estimates of a commonpopulationmean.
In random samplesof size n, the sampleproportionp fluctuates aroundthe proportionmean: rc proportion with a proportionvariance of I#9 standarderror of ,[Wdf; As the sampling distribution of p increases, it more aroundits targetmean. It also concentrates sets closerto the normal distribution.In which
iooo. wqov' z: Pft ^Tn(ttyn
Top row=.05,Bottom row=.01 points for distribution of F
of freedom for numerator
ISBN t 5?eeaEqlg
(
z
tlltt ilnfift ,llilJllflill[[llllillll
7
= mean number of runs n, : number of outcomes of one type n2 = number of outcomes of the other type S4 = standard deviation of the distribution of the numDer oI runs.
quickstudy.com
lllllilllilffillll[l
C u s t o m e rH o t l i n e# 1 . 8 0 0 . 2 3 0 . 9 5 2 2
5 . 0 6  1 . 6 1o r 5.ll.l.1ll \ l r n u l l i e t u r e cb l r I l n r i n a t r n gS e r rr c e s .I n c . l . a r g o .f l . i r n d s o l d u n d e r l i c e n s et f o r f I . r . 1 floldrng:.LL(..Prlo\lto.(.\lS.{underoneornroreolthetbllo$ingPa ts :rr S nP Lt e rr
U .5.$4.95 / C A N .$7. 50
Febr uar y 2003
\Oll('EIOSTIDENT:ThisQ{/(/t chartcorersthebasrcsofIntroduer,,r' g r r i r l r , l r r dn o l a \ a r ( p l a ( r trstics. D u e t o i 1 :c o n d c n \ c d l o r n i a t .h c t u e r cr.uscit rs r Statistics nrrnt lor assigned course work.t l0(ll. tsar( hart\. lnc. Boca Raton. FI
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.