Class Notes (836,680)
Canada (509,870)
York University (35,328)
Psychology (4,108)
PSYC 1010 (1,345)
Lecture

2-PSY1010_Lecture_3.pdf

6 Pages
95 Views
Unlock Document

Department
Psychology
Course
PSYC 1010
Professor
Rebecca Jubis
Semester
Fall

Description
PSY1010  Section  K   LECTURE  3  September  20  2012         GENERAL  ANNOUNCMENTS   -­‐ check  moodle,  weekly  lecture  notes  and  weekly  exercises  are  up.  They  might   not  be  for  every  week   -­‐ textbook  has  great  study  guide  for  practice  questions  etc   -­‐ psych  portal  “psychsim”  which  covers  topics  module  by  module  and  excellent   to  further  understanding  of  class  material.   -­‐ Left  off  last  week  talking  about  placebo  and  control  groups  –  GOOD  SHORT   ANSWER  QUESTION  ON  UPCOMING  TEST   EXPERIMENTAL  METHODS  (continued  from  last  week)   -­‐ Independent  and  dependent  variables:  have  to  have  these  in  experiments   -­‐ The  confounding  variable  is  what  you  don’t  want  in  your  experiment   o Eg  if  interested  in  looking  at  effect  of  study  times  and  what  effect  that   has  on  test  grade.  We  have  3  study  time  groups  and  average  grade   from  each  group.  Prediction  is  that  more  you  study  the  higher  the   grade.  IV  is  study  time  (manipulated)  and  DV  is  study  grade   (behavior).  Control  group  is  zero  hours  of  study  ie  group  that  does  not   study  or  doesn’t  get  IV.     o Results  look  like:    0  hours  =  65%    5  hours  =  73%    10  hours  =  65%   o how  can  people  who  study  0  hours  get  same  grade  as  those  who  study   10  hours.   o We  talk  to  participants  who  studied  10  hours  and  find  out  all  these   people  stayed  up  entire  night  before.  AHA!  Maybe  that  is  responsible   for  lower  grade.  Sleep  deprivation  might  be  influencing  grade  because   affects  performance.     o So  the  0  hour  and  5  hour  group  got  a  typical  nights  sleep   o Important  that  groups  are  equal  in  every  possible  except  for  the  IV,   which  is  the  manipulation.   o Here  the  groups  are  ALSO  different  in  hour  of  sleep  prior  to  exam   o Hence,  amount  of  sleep  is  a  CONFOUNDING  VARIABLE   o Want  to  say  iv  DV,  however  CV  can  interfere  with  this.  Cannot   determine  causation  anymore.   o Definition:  a  variable  that  interferes  with  the  results  of  your  study.  It   affects  the  DV  so  you  don’t  know  whether  the  effect  is  caused  by  IV,   the  CV  or  by  both.     o one  solution  to  this  study  is  introducing  sleep  as  another  IV.  Could   split  each  group  into  two  IV  levels.  So  now  examining  whether  sleep   effects  grade  AND  also  how  study  time  also  effects  grades.  Or  you  can   just  stress  to  the  groups  to  sleep  the  regular  amount  of  time.     1   PSY1010  Section  K   LECTURE  3  September  20  2012       o How  to  make  sure  groups  are  all  equal:  RANDOM  ASSIGNMENT  TO   GROUPS.      By  using  this  you  are  assuming  that  all  individuals  difference   variables  are  being  evenly  distributed  among  groups  so  that   the  groups  are  the  same.  “Individual  difference”  variables   mean  just  how  generally  people  differ  from  one  another.  This   is  NOT  the  same  thing  as  RANDOM  SAMPLING.      RANDOM  SAMPLING  is  when  you  do  a  survey  and  you  want  it   to  represent  the  general  population,  you  want  it  to  randomly   pick  out  recipients.     o How  would  you  study  effects  of  child  abuse  on  child  self  esteem  using   these  experimental  methods?    CANT  DO  AN  EXPERIMENT  with  this    Use  a  different  method  of  research:    CORRELATIONAL   METHOD     -­‐ CORRELATIONAL  METHOD    This  method  looks  are  relationship  between  two  variables  or   do  they  vary  together.    It  measures  the  degree  of  relationship  between  two  variables.   There  is  no  attempt  to  manipulate  or  control  variables.  Rather   naturally  occurring  changes  on  both  variables  are  measured  to   see  if  they  are  related.  Correlations  also  help  us  make   predictions.      So  if  you  wanted  to  study  effect  of  fatigue  on  study  grade,  To   measure  the  correlation  we  could  get  a  rating  of  fatigue  and   their  grade.  So  have  two  scores  for  each  person.  Then  we  plot   these  onto  a  scatter  plot  using  fatigue  rating  as  x  value  and   grade  as  y  value.      If  variables  related,  we  observe  a  linear  trend  which  means   they  are  correlated.    If  say  looked  at  GPA  and  shoe  size  correlation,  points  on  scatter   plot  would  be  all  over  the  place  and  probably  not  correlated.   Knowing  shoe  size  does  not  let  us  predict  GPA.    Correlation  value  is  called  correlation  coefficient   represented  by  “r”  and  can  be  between  -­1  and  1.  It  can   NEVER  be  more  than  +  or  –  1.      Value  of  number  tells  strength  of  relationship  between  two   variables    Negative  correlation  means  that  as  one  variable  increases  in   value,  the  second  variable  decreases  in  value.  This  is  between  -­‐ 1  and  0.  If  r=  -­‐0.9,  then  we  know  that  these  two  variables  are   STRONGLY  negatively  correlated.  So  for  eg  if  you  look  at   correlation  between  GPA  and  hours  spent  watching  TV,  we     2   PSY1010  Section  K   LECTURE  3  September  20  2012       would  expect  a  negative  correlation.  So  as  the  number  of  hours   spent  watching  TV  increase,  the  GPA  should  decrease.  So  the   value  would  be  negative  and  the  actual  number  would  tell  the   strength  of  this  relationship    Positive  correlation  means  that  as  one  variable  increases  in   value,  the  second  variable  also  increases  in  value.  This  value   will  always  be  between  0  and  +1.  So  if  for  e.g.  r=0.75,  we  know   that  two  variables  are  both  increasing  in  the  same  direction.   Now,  the  values  can  BOTH  be  either  direction  both  values  can   either  increase  together  or  decrease  together.  So  for  e.g.  now   looking  at  hours  of  study  and  GPA.  As  hours  of  study  increases,   GPA  also  increases.  Also  can  say  as  hours  of  study  decreases,   the  GPA  also  decreases.      So  going  back  to  child  abuse  and  self  esteem,  if  get  a  correlation   value  of  -­‐0.021.     • Value  of  number  ie  0.021  tells  us  that  it  a  WEAK   correlation   • Sign  (-­‐)  tells  us  that  it  is  a  positive  correlation  and  that   as  child  abuse  score  increases,  self  esteem  decreases,   even  if  the  relationship  is  weak.      IMPORTANT:  CORRELATION  DOES  NOT  MEAN  CAUSATION.   So  if  two  variables  are  increasing  together,  that  DOES  NOT   mean  that  one  variable  is  causing  the  second  variable  to   increase.    All  you  can  say  is  that  these  two  variables  are   related.  For  e.g.  there  is  a  strong  positive  relaionship  between   amount  of  exercise  and  psychological  well  being.  There  is  a   temptation  to  say  that  exercise  causes  psychological  well   being.  However,  we  don’t  know  which  one  c
More Less

Related notes for PSYC 1010

Log In


OR

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


OR

By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.


Submit