Statistical Analysis, Chapter 4

Overview  
•  Preparing  data  for  sta2s2cal  analysis  
•  Descrip2ve  sta2s2cs  
•  Comparing  means  
•  T  tests  
•  Analysis  of  variance  (ANOVA)  
•  Assump2ons  of  t  tests  and  F  tests  
•  Iden2fying  rela2onships  
•  Regression  
•  Nonparametric  sta2s2cal  tests  
 Preparing  data  for  analysis  
•  Cleaning  up  data  
–  Detect  errors  
–  FormaCng  
•  Coding    
–  Types  of  data  that  need  to  be  coded  
–  Be  consistent  
•  Organizing  the  data  
–  Accommodate  to  the  requirements  of  sta2s2cal  soIware  
 Descrip2ve  sta2s2cs  
•  Measures  of  central  tendency  
–  Mean  
–  Median  
–  Mode  
•  Measures  of  spread  
–  Range  
–  Variance  
–  Standard  devia2ons  
 
 Videos  

http://www.khanacademy.org/math/statistics/v/small-sample-hypothesis-test
http://www.youtube.com/watch?v=leHOBf_-9kM
 Comparing  means  
•  Summary  of  methods  
 Comparing  2  means:  T  tests  
•  Independent-­‐samples  t  test:  between-­‐group  design  
 Comparing  2  means:  T  tests  
•  Paired-­‐sample  t  test:  within-­‐group  design  
 Comparing  2  or  more  means:  
Analysis  of  variance  (ANOVA)  
•  Also  called  F  tests  
•  One-­‐way  ANOVA:  for  between-­‐group  design  
•  Data  layout:  Table    4.6  
•  Results  summary:  
 Factorial  ANOVA  
•  For  between-­‐group  design  
•  2  or  more  independent  variables  involved  
•  Data  layout:  table  4.9  
 Factorial  ANOVA  
•  Summary  results  
 Repeated  measures  ANOVA  
•  For  within-­‐group  design  
•  Can  inves2gate  one  or  more  variables  
•  One-­‐way  ANOVA  
 Repeated  measures  ANOVA  
•  One  way  ANOVA  summary  report:  
 Repeated  measures  ANOVA  
•  Two  way  ANOVA  experiment  design:  
 Repeated  measures  ANOVA  

Two  way  ANOVA  data  layout  

 Repeated  measures  ANOVA  
•  Two  way  ANOVA  summary  report:  
 Split-­‐plot  ANOVA  
•  Involves  both  between-­‐group  and  within-­‐group  
factors  
•  Experiment  design  
Split-­‐plot  ANOVA  data  layout  
Split-­‐plot  ANOVA  summary  report  
Interpre2ng  test  sta2s2cs,  p-­‐values  
 1) The null hypothesis here is that the means are equal, and the alternative
hypothesis is that they are not. A big t, with a small p-value, means that the
null hypothesis is discredited, and we would assert that the means are
significantly different (while a small t, with a big p-value indicates that they are
not significantly different).
2) The null hypothesis here is that one mean is greater than the other, and the
alternative hypothesis is that it isn't. A big t, with a small p-value, means that
the null hypothesis is discredited, and we would assert that the means are
significantly different in the way specified by the null hypothesis (and a small t,
with a big p-value means they are not significantly different in the way
specified by the null hypothesis).
3) The null hypothsis here is that the group means are all equal, and the
alternative hypothesis is that they are not. A big F, with a small p-value,
means that the null hypothesis is discredited, and we would assert that the
means are significantly different (while a small F, with a big p-value indicates
that they are not significantly different).
 4) The null hypothsis here is that the group variances are all equal, and the
alternative hypothesis is that they are not. A big K2, with a small p-value, means that
the null hypothesis is discredited, and we would assert that the group variances are
significantly different (while a small K2, with a big p-value indicates that they are not
significantly different).
5) The null hypothesis here is that there is not a general relationship between the
response (dependent) variable and one or more of the predictor (independent)
variables, and the alternative hypothesis is that there is one. A big F, with a small p-
value, means that the null hypothesis is discredited, and we would assert that there is
a general relationship between the response and predictors (while a small F, with a
big p-value indicates that there is no relationship).
6) The null hypothesis is that the value of the p-th regression coefficient is 0, and the
alternative hypothesis is that it isn't. A big t, with a small p-value, means that the null
hypothesis is discredited, and we would assert that the regression coefficient is not 0
(and a small t, with a big p-value indicates that it is not significantly different from 0).
 Iden2fy  rela2onships  
•  Correla2on:  Two  factors  are  correlated  if  there  is  a  
rela2onship  between  them  
•  Most  commonly  used  test  for  correla2on  is  the  
Pearson’s  product  moment  correla2on  coefficient  
test  
•  Pearson’s  r:  ranges  between  -­‐1  to  1  
•  Pearson’s  r  square  represents  the  propor2on  of  the  
variance  shared  by  the  two  variables  
 Iden2fy  rela2onships  
•  Correla2on  does  not  imply  causal  rela2onship  
 Iden2fy  rela2onships  
•  Regression:  can  inves2gate  the  rela2onship  
between  one  DV  and  mul2ple  IVs  
•  Regression  is  used  for  2  purposes:  
–  Model  construc2on  
–  Predic2on  
•  Different  regression  procedures  
–  Simultaneous  
–  Hierarchical  
 
 Non-­‐parametric  tests  

•  Non-­‐parametric  tests  are  used  when:  

–  The  error  is  not  normally  distributed  
–  The  distances  between  any  two  data  units  are  not  
equal  
–  The  variance  of  error  is  not  equal  
 Non-­‐parametric  tests  
•  CHI-­‐square  test  
–  Used  to  analyze  categorical  data  
–  Table  of  counts  (con2ngency  table)  
–  Assump2ons  of  the  test  
•  Data  points  need  to  be  independent    
•  The  sample  size  should  not  be  too  small  
 Non-­‐parametric  tests  
•  Two  groups  of  data  
–  For  between-­‐group  design:  Mann–Whitney  U  test  
or  the  Wald–Wolfowitz  runs  test  
–  For  within-­‐group  design:  Wilcoxon  signed  ranks  
test  
•  Three  or  more  groups  of  data  
–  For  between-­‐group  design:  Kruskal–Wallis  one-­‐
way  analysis  of  variance  by  ranks  
–  For  within-­‐group  design:  Friedman’s  two-­‐way  
analysis  of  variance  test