Your SlideShare is downloading. ×

Thanks for flagging this SlideShare!

Oops! An error has occurred.

×

Introducing the official SlideShare app

Stunning, full-screen experience for iPhone and Android

Text the download link to your phone

Standard text messaging rates apply

How to Choose a Sample for Your Thesis or Dissertation

640
views

Published on

Concepts, relationships, and types of samples are presented. Cases to practice sampling are included. Sample size is also discussed.

Concepts, relationships, and types of samples are presented. Cases to practice sampling are included. Sample size is also discussed.

Published in: Education, Technology, Business

0 Comments
3 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total Views
640
On Slideshare
0
From Embeds
0
Number of Embeds
0
Actions
Shares
0
Downloads
7
Comments
0
Likes
3
Embeds 0
No embeds

Report content
Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
No notes for slide

Transcript

  • 1. How  to  Choose  a   Sample  for  Your  Thesis   or  Dissertation  
  • 2. Concepts  to  Know   Research  Popula.on   The  whole  set  of  units  (people,  groups,  events,   se4ngs,  etc.)  on  which  the  research  is  focused  and   the  findings  are  expected  to  be  generalized.   Sample   Representa?ve  subset  of  the  popula?on  the   researcher  studies  when  the  whole  popula?on  is  not   available     Sampling  Scheme   Specific  strategies  to  select  samples   Sampling  Design   Umbrella  term  that  includes  both  selec?ng  strategies   and  sample  size  
  • 3. Relationships  to   Remember   Research   Problem   •  What  needs  to  be  studied  under  a  specific   situa?on   Research   Ques?on   •  What  to  answer  or  test   Methodology   •  How  to  gather  and   analyze  valid  data  
  • 4.   Sampling   Design   Research   Design   Data   Analysis   Methods   Valid  Informa.on  to  Answer  or  Test   Research  Ques.on  
  • 5. The  next  slides  will  present  a  basic  summary   of  some  of  the  sampling  schemes  that  are   widely  used  in  research.  However,  it  is   necessary  to  say  that  the  applica?on  of   probabilis?c  (random)  schemes  to   quan?ta?ve  research  and  non-­‐probabilis?c   (non-­‐random)  ones  to  qualita?ve  studies  does   not  imply  that  these  are  the  only  approaches.    
  • 6. In  fact,  Onwuegbuzie  &  Collins  (2007)  state  that  some  form  of   non-­‐random  scheme  is  the  most  common  choice  used  in  both   quan?ta?ve  and  qualita?ve  studies  and  that  random  ones  for   quan?ta?ve  research  and  non-­‐random  for  qualita?ve  studies   is  the  second  most  common  combina?on.      Our  own  research  into  disserta?on  and  thesis  work   interna?onally  demonstrates  that  the  majority  employ   purposeful  convenience  samples.    Such  predominance  of   non-­‐random  samples  is  related  to  the  fact  that  most  studies   in  social  sciences  cannot  be  done  under  experimental   condi?ons  in  which  pure  random  selec?on  is  expected.     Onwuegbuzie,  A.  &  Collins,  K.  (2007).  A  typology  of  mixed  methods  sampling   designs  in  social  science  research.  The  Qualita+ve  Report,  12  (2),  281-­‐316      
  • 7. Quantitative  Research   Simple  random     The  whole  popula?on  is  available  and  any  unit  has  the   same  chance  of  being  chosen     Stra.fied  random   The  whole  popula?on  is  divided  into  subpopula?ons   (strata)  with  respect  to  one  or  more  characteris?cs  that   interest  and  units  are  selected  from  each  stratum  at   random.  Alloca?on  of  units  can  be  done  equal  or   propor?onal  to  the  popula?on     Systema.c   Popula?on  is  ordered  according  to  a  criterion  and  units  are   chosen  from  the  list  by  selec?ng  every  nth  one     Cluster   Popula?on  consists  of  limited  groups  (clusters)  and   sampling  is  focused  on  selec?ng  clusters  instead  of   individual  units.  Not  all  clusters  are  included  in  the  sample     Mul.stage  random   Very  large  popula?ons  are  divided  into  clusters  and  then   sub-­‐clusters  and  units  are  selected  at  random  following  a   general  to  specific  direc?on    
  • 8. Qualitative  Research  (I)   Convenience   Sample  consists  of  units  that  are  available  and/  or   willing  to  par?cipate     Purposeful   Researcher  is  interested  in  studying  specific  groups.   Selec?on  of  units  can  be  done  at  random,  stra?fied,   or  using  more  than  one  scheme  (mixed)     Quota   Units  are  selected  in  rela?on  to  pre-­‐defined   characteris?cs  either  in  propor?on  to  popula?on   sub-­‐groups  or  minimum  number  from  each  sub-­‐ group     Snowball   Similar  units  are  required  and  access  to  them  is  done   by  asking  par?cipants  to  recommend  peers     Mul.stage  purposeful     Units  are  selected  in  more  than  one  stage  and   always  applying  a  purposive  scheme    
  • 9. Qualitative  Research  (II)   Typical  case   Units  are  chosen  because  they  represent  the   average  element  of  what  is  studied     Maximum  varia.on   When  differences  are  the  research  target,  dissimilar   units  are  chosen  to  form  sample     Criterion   Units  are  selected  because  each  one  of  them   represents  one  or  more  desired  criteria     Theore.cal   Units  are  chosen  because  they  can  provide  input   informa?on  to  build  or  test  a  theory    
  • 10. Mixed  Methods  Research   All  previous  schemes   When  choosing  samples  for  mixed  methods  studies,   researchers  need  to  take  into  account  the  purpose  of   the  study,  the  research  ques?ons,  and  the  stages  in   which  the  study  will  be  developed.  All  those   elements  help  them  select  the  most  appropriate   sampling  scheme  for  each  stage.  When   generaliza?on  is  the  main  concern  of  a  stage,  the   first  five  schemes  presented  are  the  best  op?ons.   When  understanding  is  the  focus  of  a  stage,  then  the   rest  of  schemes  should  be  considered.    
  • 11. Cases  to  Practice  Sampling  
  • 12. Quantitative  Research   • The research director of an educational city system that includes forty-five elementary schools wants to answer the following research question: What’s the achievement in mathematics of fifthgrade students of the school system, measured by a standardized achievement test? Options for Sampling: 1.  Simple random 2.  Stratified random 3.  Cluster And the answer is…….
  • 13. Cluster  Sampling   It’s  too  expensive  to  administer  the  test  to  the  whole  popula?on  and  the  same  happens  if  the   simple  random  sampling  is  chosen.  Stra?fied  random  sampling  implies  administering  the  test  to   some  of  the  students  of  a  class  and  not  to  others  and  that  may  be  inconvenient  from  the   students’  point  of  view.  Since  students  are  organized  in  classes  –which  are  in  fact  clusters,  the   most  appropriate  op?on  is  cluster  sampling  and  that  will  allow  to  test  all  students  in  the  chosen   classes.    
  • 14. Qualitative  Research   •  A researcher wants to know the reasons why some employees of large companies have strong pro-mentoring duties views. Options for Sampling: 1.  Purposeful 2.  Snowball 3.  Maximum variation And the answer is…….
  • 15. Snowball   Since  par?cipants  with  similar  views  within  organiza?ons  are  required  and  they  could  be  difficult   to  locate,  it  is  appropriate  to  ask  them  to  recommend  poten?al  subjects.        
  • 16. Sample  Size  
  • 17. Criteria  to  Select  Sample  Size  (I)   •  Costs  (money,  ?me,  and  effort)  to  get  sample  data.   •  For  quan?ta?ve  studies:     –  Popula?on  size  (the  larger  the  popula?on,  the  larger  the   sample)   –  Confidence  interval  (how  much  error  will  be  allowed)   –  Confidence  level  (how  much  confident  you  want  to  be  that   your  results  are  within  the  selected  confidence  interval)   –  Standard  devia?on  (how  much  dispersion  from  the  mean  you   expect)   These  criteria  is  used  by  calculators  to  determine  the  most   appropriate  sample  size  for  your  study  
  • 18. Criteria  to  Select  Sample  Size   (II)     •  For  qualita?ve  studies:      Samples  are  usually  small,  but  the  precise  number  is  very   unlikely  to  be  determined  at  the  beginning  of  the  study  .   The  main  criterion  is  to  have  a  sample  as  big  as  needed   to  have  all  the  informa?on  that  might  be  important.   Therefore,  when  informa?on  becomes  redundant,  the   sample  size  has  been  reached.  
  • 19. Criteria  to  Select  Sample  Size   (III)   •  For  mixed  studies:     –  Usually  the  sizes  of  the  sample  for  each  stage  are  different   –  The  mixed  methods  variant  used  influences  the  sample  sizes.   For  example,  in  an  explanatory  design  the  same  par?cipants   must  be  included  in  all  the  stages  while  in  an  exploratory   design  more  par?cipants  are  needed  in  the  quan?ta?ve  stage.    
  • 20. Now  You  Know   1.  The relationship between research problem, research question, and sampling. 2.  Some of the basic sampling schemes you may use to select participants for your study Any  question?  
  • 21. Thanks  for  Coming   Hope you find this conference useful and want to meet us soon