2.

1  Worksheet  -­‐  Developing  a  Thesis  /  Starting  ISU  Project  

MDM4U  
Jensen    
 
Your  first  task  is  to  identify  a  topic  for  your  project.  Your  topic  needs  to  be  of  interest  to  you  personally.  It  
must  also  involve  sufficient  mathematical  content  and  analysis  to  constitute  a  reasonable  culminating  
project.    
 
Some  possible  sources  for  ideas  include:  
 
• A  sport  or  hobby  that  you  enjoy  
• A  social  issue  of  interest  to  you  
• An  interesting  issue  from  the  Internet  or  other  media  
• An  issue  arising  from  employment  or  a  possible  future  career  
 
When  determining  your  topic,  it  is  important  to  note  the  difference  between  a  subject  and  a  topic.  The  
word  subject  denotes  a  broad  category,  a  general  area  of  interest  that  you  may  wish  to  explore  further.  
The  word  topic  denotes  a  specific  aspect  of  the  subject  that  requires  detailed  research.  Complete  the  
chart  below  before  choosing  your  topic.  Add  one  topic  to  each  subject  and  create  your  own  example.  
 
Subject   Topic  
Television   Ratings  wars  
Digital  cable  versus  satellite    
 
 
Medicine   Hospital  closures  and  staff  cutbacks  
Doctor/nurse  shortages  
Health-­‐care  funding  levels  
 
 
Post-­‐secondary  education   Tuition  fees  
Entry  requirements  
Degree  requirements  
 
 
Travel   Air  safety  
Vacation  destinations  
Cost  
 
 
Music   Musical  styles  
Music  and  sports  
Comparing  musicians  
 
 
   
 
 
 
 
A  useful  tool  for  generating  and  organizing  related  topics  is  called  a  mind  map.  A  mind  map  begins  with  a  
broad,  general  subject  and  generates  topics  and  sub-­‐topics  related  to  the  main  subject.    
 
Mind  maps  typically  have  several  levels.  Each  level  flows  from  the  level  above  it.  The  magnitude  of  the  
topics  at  each  level  is  smaller  than  the  previous  level.    
 
Select  three  of  the  following  subjects:  automobiles,  music,  movies,  sports,  dance,  fashion,  travel,  
occupations,  crime.  Then,  brainstorm  topics  and  sub-­‐topics  and  construct  a  mind  map  for  each  of  your  
three  topics.  Try  to  include  several  levels  in  your  mind  maps.    
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Once  you  have  narrowed  down  your  topic,  you  will  have  to  create  a  well-­‐written  thesis  question  or  
statement  that  clarifies  exactly  what  your  project  is  designed  to  do.  Your  project  must  aim  to  answer  a  
specific  problem.  The  problem  should  satisfy  all  of  the  following:  
 
o Significant  problem  of  interest  to  YOU  
o Involve  the  collection  of  a  large  amount  of  data  
o Involve  the  organization  of  a  large  amount  of  data  
o Involve  the  analysis  of  a  large  amount  of  data  
o Allow  the  use  of  technology  
 
 
When  developing  your  thesis  question,  use  the  following  series  of  questions  to  determine  if  your  thesis  
question  is  effective:  
 
1) Does  the  thesis  question  clearly  identify  the  main  objective  of  my  presentation?  
2) Does  my  thesis  question  clearly  identify  the  main  variables  that  will  be  studied?  
3) Does  my  thesis  question  use  specific  language  that  tells  audience  what  I  am  going  to  do,  rather  
than  vague  language?  
4) Is  my  thesis  question  interesting?  Does  it  make  me  want  to  research  further?  
5) Is  my  thesis  question  too  general  or  too  specific?  
6) Does  my  thesis  question  show  evidence  of  original  thought?  Is  the  topic  I  am  proposing  worth  
writing  about?  
 
 
Example:  Suppose  you  plan  to  investigate  the  effect  of  increased  tuition  fees  on  university  accessibility.  
Two  possible  thesis  statements/questions  you  could  use  to  investigate  this  topic  are:  
 
1) Increases  in  tuition  fees  over  the  last  ten  years  have  forced  large  numbers  of  middle  class  students  
to  give  up  on  a  university  education.  
2) Increases  in  tuition  fees  over  the  last  ten  years  have  allowed  universities  to  broaden  the  range  of  
specialty  programs  offered.    
 
Notice  that  in  the  examples  you  are  not  initially  stating  that  your  hypothesis  is  true.  You  are  taking  a  
position,  which  you  will  test  by  collecting  and  analyzing  data.    
 
Example:  Gord  loves  to  watch  TV  and  would  like  to  study  the  TV-­‐watching  habits  of  his  classmates.  
Create  three  thesis  questions  related  to  this  topic  that  can  be  statistically  studied.  
 
1)    
 
 
 
2)  
 
 
 
3)  
 
 
 
 
  
Example:  Determine  the  main  variables  in  the  following  thesis  questions  
 
a) Are  females  better  than  males  at  estimating  the  size  of  large  crowds?  
 
The  main  variables  are:  gender,  estimates  of  size  of  crowds,  actual  size  of  crowds  
 
 
b) Is  there  a  relationship  between  the  quality  of  a  person’s  clothing  and  her  or  his  mid-­‐term  average?  
 
The  main  variables  are:    
 
 
 
 
 
Example:  For  each  pair  of  variables,  create  a  thesis  statement/question  that  could  be  tested.  
 
1) Car  insurance  and  driver  age  
 
Thesis:  Insurance  premiums  for  young  drivers  are  higher  because  they  are  at  greater  risk  of  causing  
an  accident.    
 
 
2) Sports  team  profits  and  player  salaries  
 
Thesis:  
 
 
 
 
 
3) Income  and  level  of  education  
 
Thesis:  
 
 
 
 
 
 
   
 
 
 
 
 
 
 
 
 
 Your  Project  
 
Task  1:  Design  a  mind-­‐map  of  your  interests  with  at  least  three  levels.    
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Task  2:  Select  a  topic  
 
The  topic  you  are  going  to  base  your  project  on  is:  _____________________________________________________________  
 
 
Task  3:  Create  a  thesis  question  
 
Develop  three  possible  thesis  questions/statements  related  to  your  topic  of  interest  
 
1)    
 
 
 
2)    
 
 
 
3)    
 
 
 
Task  4:  Conduct  preliminary  data  search  on  your  topic.  The  type  and  quantity  of  data  available  may  
require  you  to  change  your  topic  or  thesis  question.  
2.2  Worksheet  Characteristics  of  Data  –  Worksheet  
MDM4U  
Jensen  
 
 
1)  Identify  each  of  the  following  variables  as  quantitative  or  qualitative.  For  each  quantitative  variable,  
identify  whether  it  is  continuous  or  discrete.    
 
a)  age    
 
quantitative,  usually  discrete  
         
b)  favourite  meal  
 
qualitative  
       
c)  television  viewing  preferences  
 
qualitative  
 
d)  speed  of  car    
 
quantitative,  continuous  
     
e)  colour  of  hair  
 
qualitative  
       
f)  fabric  texture  
 
qualitative  
 
g)  pH  of  water  samples  
 
quantitative,  continuous  
     
h)  seating  capacity  
 
quantitative,  discrete  
       
i)  test  mark  
 
quantitative,  discrete  
 
j)  paint  colours  
 
qualitative  
 
 
2)  Identify  the  variables  and  their  types,  as  well  as  the  population  for  the  following  thesis  questions.  Also,  
would  you  collect  a  sample  or  conduct  a  census?  Would  each  question  require  a  cross-­‐sectional  study  or  a  
longitudinal  study?  
 
a)  Is  there  a  relationship  between  weather  conditions  and  absenteeism  in  Grade  9  at  your  school?  
 
Weather  condition  is  a  qualitative  variable  (can  be  quantitative  and  continuous  if  looking  at  temperature).  
Absenteeism  is  a  quantitative  and  discrete.  The  population  is  grade  9  students  in  our  school.  Sample  is  collected.  
Longitudinal  study  would  be  required  to  track  attendance  records  over  a  period  of  time  with  different  weather  
conditions.    
 
b)  Is  there  a  relationship  between  the  amount  of  television  watched  and  the  level  of  physical  fitness  
among  adult  females?  
 
Amount  of  television  is  a  quantitative  and  continuous  variable  (measured  in  minutes).  Physical  fitness  is  a  
quantitative  and  continuous  variable  (using  BMI).  The  population  is  adult  females.  Sample  is  collected.  Cross  sectional  
study  would  be  easiest  to  do  but  longitudinal  is  an  option.  
 
 
c)  Are  teenage  drivers  who  have  been  issued  speeding  tickets  more  likely  to  be  males?  
 
Gender  is  a  qualitative  variable.  Number  of  female  students  with  speeding  tickets  is  a  quantitative  and  discrete  
variable.  Number  of  male  students  with  speeding  tickets  is  a  quantitative  and  discrete  variable.  The  population  is  
teenagers  who  have  been  issued  speeding  tickets.  Sample  is  collected.  Cross-­‐sectional  study  is  required.  
 
 
3)  Consider  this  thesis  question:  In  North  America,  do  foreign  cars  depreciate  in  value  faster  than  domestic  
cars?  Now  answer  the  questions  that  follow:  
 
a)  What  is  the  population?  
 
The  population  is  cars  in  North  America.  
 
b)  What  are  the  key  variables  that  must  be  considered?  Are  these  quantitative  or  qualitative?  If  
quantitative,  are  they  discrete  or  continuous?  
 
Type  of  car  is  a  qualitative  variable.  Value  of  car  is  a  quantitative  discrete  variable.    
 
c)  Should  a  census  or  a  sample  be  used  to  collect  data?  
 
A  sample  should  be  used.  
 
d)  Is  a  cross-­‐sectional  or  a  longitudinal  study  more  appropriate  for  drawing  conclusions?  
 
Cross-­‐sectional  
 
 
 
 
 
 
 
4)  Explain  the  differences  between  each  pair  of  terms.    
 
a)  population/sample  
 
Population  is  the  group  being  studied  when  sample  is  a  selection  of  individual  taken  from  the  population.  
           
b)  cross-­‐sectional  study/longitudinal  study  
 
Cross-­‐sectional  study  is  a  study  that  considers  individuals  from  different  groups  at  the  same  time.  
Longitudinal  study  is  a  study  of  a  single  group  (or  sample)  over  a  long  period  of  time.  
 
c)  quantitative  variable/qualitative  variable  
 
Quantitative  data  are  numerical  and  qualitative  data  are  non-­‐numerical.  
     
d)  discrete  data/continuous  data  
 
Discrete  data  is  data  that  can  only  take  on  a  finite  number  of  values  within  a  given  range.  For  example,  number  of  
vehicles  is  a  discrete  data.  Continuous  data  is  data  that  are  measurable  with  all  real  numbers  and  therefore  can  take  
on  an  infinite  number  of  values  within  a  given  range.    
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2.3  Worksheet  -­‐  Collecting  Samples  
MDM4U  
Jensen  
 
1)  Identify  the  type  of  random  sampling  in  each  of  the  following  scenarios.  
 
a) The  principal  randomly  selects  four  classes  and  surveys  each  student  in  those  classes  
 
Cluster  random  sampling  
 
b) William  picks  names  out  of  a  hat  
 
Simple  random  sampling  
 
c) A  hockey  card  collector  opens  a  drawer  of  sorted  cards  and,  after  selecting  a  random  starting  
point,  takes  out  every  fifth  card.    
 
Systematic  random  sampling  
 
d) The  Ministry  of  Education  randomly  selects  your  school  for  testing,  and  40  student  names  are  
randomly  selected  from  a  student  list.    
 
Multi-­‐stage  random  sampling  
 
e) Your  class  submits  solutions  to  a  problem  and  your  teacher  divides  the  work  into  four  piles  by  
achievement  levels.  She  then  randomly  picks  three  examples  from  each.    
 
Stratified  random  sampling  
 
f) A  farmer  brings  a  juice  company  several  crates  of  oranges  each  week.  A  company  inspector  looks  
at  10  oranges  from  the  top  of  each  crate  before  deciding  whether  to  buy  all  the  oranges.  
 
Convenience  non-­‐random  sampling  
 
g) The  ABC  program  Nightline  once  asked  whether  the  United  Nations  should  continue  to  have  its  
headquarters  in  the  United  States.  Viewers  were  invited  to  call  one  telephone  number  to  respond  
‘yes’  and  another  for  ‘no.’  More  than  186  000  callers  responded.    
 
 
Voluntary  non-­‐random  sampling  
 
 
 
 
 
 
 
 
 
 
2)  A  textbook  has  600  pages  and  6  chapters.  Describe  how  to  you  could  design  and  carry  out  the  
following  samples  of  its  pages.    
 
a) Select  6  pages  from  the  textbook  using  simple  random  sampling  
 
Use  randint(1,  600,  6)  to  randomly  select  6  pages.  
 
b) Select  10  pages  using  systematic  random  sampling  
 
select  random  starting  point  using  randint(1,  600,  1)  and  then  select  every  60th  page  (sampling  interval  =  600/10  
=  60)  
 
c) Select  12  pages  using  stratified  random  sampling  
 
Divide pages into groups based on chapters. Take a simple random sample of 2% (12/600 = 0.02) of the pages from
each chapter.
 
d) Select  10  pages  using  multi-­‐stage  random  sampling.    
 
Divide pages into groups based on chapter. Do a simple random sample of chapters and then do a simple random
sample of the pages within the chosen chapters.  
 
 
3)  Based  on  the  following  groups  of  names,  identify  a  sampling  method  that  may  have  been  used  to  
collect  the  samples  listed  in  parts  (a)  through  (e).  
 
Shaggy     Paul       Joey       Susan  
Fred       John       Monica     Elmo  
Scooby     George     Rachel       Ernie  
Thelma     Ringo       Ross       Oscar  
Daphne           Chandler     Zoe  
            Phoebe     Maria  
 
 
a) Joey,  Monica,  Fred,  Paul,  Daphne           SIMPLE  
b) Susan,  Elmo,  Ernie,  Oscar,  Zoe,  Maria         CLUSTER  
c) Shaggy  Scooby,  Daphne             SYSTEMATIC  
d) John,  George,  Ringo               MULTI-­‐STAGE        
e) Shaggy,  Fred,  George,  John,  Joey,  Chandler,  Susan,  Ernie     STRATIFIED  
 
 
 
 
 
 
 
 
 
 
 
 
 
4)  Student  council  wants  to  conduct  a  survey  during  the  first  five  minutes  of  an  assembly.    There  are  800  
students  at  the  assembly.  A  map  of  the  auditorium  is  shown  below.  Note  that  the  students  are  seated  by  
grade  level  and  the  seats  are  numbered  from  1  to  800.  Describe  how  you  would  use  your  calculatr  to  
select  80  students  to  complete  the  survey  with  each  of  the  following  methods:  
 
a) Simple  Random  Sample  
 
Use  randint(1,  800,  80)  to  choose  which  students  
to  give  the  survey  to.    
 
b) Stratified  Random  Sample  
 
Use  the  grade  levels  at  the  strata.  Within  each  
grade’s  seating  area,  we’ll  select  10%  (20)  of  the  
seats.  
For  9th  grade  use  randint(601,  800,  20)  
For  10th  grade  use  randint(401,  600,  20)  
For  11th  grade  use  randint(201,  400,  20)  
For  12th  grade  use  randint(1,  200,  20)  
 
 
c) Cluster  Random  Sample  
 
When  using  cluster  random  sampling,  it  is  best  if  each  cluster  has  the  same  characteristics  as  the  
population.  For  this  reason,  it  would  be  best  to  use  each  column  of  seats  as  a  cluster  because  that  will  
ensure  there  are  students  of  each  grade  level  in  each  cluster.  Because  there  are  20  columns  (clusters),  
each  with  40  seats,  we  need  to  randomly  choose  2  clusters  to  get  our  sample  of  80.  Use  randint(1,  20,  2)  
to  select  two  clusters  and  then  give  the  surveys  to  ALL  of  the  students  in  those  clusters.    
 
d) Systematic  Random  Sample  
 
Use  randint(1,  800,  1)  to  determine  a  random  starting  point.  Then  give  the  survey  to  every  10th  student  
(sampling  interval  =  800/80).    
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2.4  Worksheet  –  Survey  Design  and  Bias  
MDM4U  
Jensen  
 
 
 
1)  For  each  of  the  following  questions,  state  if  it  is  an  open  question,  rating    question,  ranking  question,  
checklist  question,  or  information  question.  
 
i)  Please  provide  the  following  information:          
 
  Gender:  _________________        Grade:  _________________  
 
INFORMATION  
 
ii)  Please  provide  the  following  information  
 
  Gender:  ☐M          ☐F                 Grade:    ☐9 ☐10 ☐11 ☐12  
 
INFORMATION  
 
iii)  With  1  meaning  most  helpful  and  10  meaning  not  at  all  helpful,  rate  each  of  the  chapters  of  the  
textbook.    
 
______  Chapter  1     ______  Chapter  2     ______  Chapter  3  
 
______  Chapter  4     ______  Chapter  5     ______  Chapter  6  
 
RATING  
 
iv)  Rank  the  chapters  of  this  textbook  in  order  from  1-­‐6  (1  being  most  useful).    
 
______  Chapter  1     ______  Chapter  2     ______  Chapter  3  
 
______  Chapter  4     ______  Chapter  5     ______  Chapter  6  
 
RANKING  
 
2)  Describe  the  characteristics  of  a  good  question  and  what  good  questions  must  avoid.    
 
Good questions are specific, simple, relevant, readable. Good questions avoid jargon, abbreviations, negatives,
being leading, and insensitivity  
 
 
 
 
 
 
 
 
3)  Using  your  criteria  from  the  previous  question,  evaluate  the  following  survey  questions  and  re-­‐write  
the  question  to  meet  all  of  the  criteria  of  a  good  question.    
 
a)  Should  the  OMB  be  funded  to  initiate  waste  audits  across  the  province?  
 
 ☐Y              ☐N              
 
abbreviation; Should  the  Ontario  Municipal  Board  be  funded  to  initiate  waste  audits  across  the  province?  ☐Y        ☐N      
 
b)  Given  the  large  amount  of  sulfur  dioxide  that  is  spewed  out  of  smelters,  should  mining  companies  be  
forced  to  clean  up  their  act?  Please  comment.    
 
leading,  insensitivity;    Should mining companies be forced to decrease the amount of sulphur dioxide being
emitted at smelters? Please comment.
 
c)  On  a  5-­‐point  scale,  do  you  agree  with  the  bleeding-­‐heart  Liberals  that  all  corporations  should  pay  
higher  taxes? (1: totally agree, 5: totally disagree)  
 
leading, insensitivity; On a 5–point scale, do you agree that all corporations should pay higher taxes? (1: totally
agree, 5: totally disagree)
 
d)  Why  shouldn’t  forestry  companies  clear  cut?  Please  explain.  
 
jargon, negatives; Should forestry companies be able to cut down all trees in certain areas? Explain.
 
 
4)  For  each  of  the  following  questions,  state  if  it  is  an  open  question,  rating    question,  ranking  question,  
checklist  question,  or  information  question.    
 
a)  You  are  presently  in  grade  (circle  the  appropriate  answer):  
 
9   10   11   12  
 
INFORMATION  
 
b)  I  find  mathematics  stimulating  because:  
 
 ________________________________________________________________________________________________________________________  
 
OPEN  
 
c)  Rank  the  following  foods  from  favourite  (1)  to  least  favourite  (4).    
 
______  pizza     ______  hamburgers     ______  tacos     ______  watermelon  
 
RANKING  
 
 
 
 
 
d)  Do  you  wear  a  wristwatch?  
 
______  always     ______  sometimes     ______  seldom     ______  never  
 
INFORMATION  
 
e)  How  much  do  you  like  math  on  a  scale  of  1  to  5  (1  being  the  lowest)  
 
RATING  
 
f)  Estimate  your  net  income:  
 
______  $15000-­‐$19999   ______$20000-­‐$39999     ______  $40000-­‐$59999   ______  $60000  +  
 
INFORMATION  
 
5)  Explain  the  difference  between  primary  and  secondary  data.    
 
If you conduct a survey and collect information firsthand, this is called primary data. This type of data is easy
to work with because you control how it is collected.

Information obtained from similar studies conducted by OTHER researchers is called secondary data.
 
 
6)  Identify  the  type(s)  of  bias  that  might  result  from  each  of  the  following  data  collection  methods.    
 
a)  You  hand  out  surveys  to  your  classmates  to  be  returned  to  you  next  week.    
 
NON-­‐RESPONSE  –  some  students  will  not  return  the  survey  
 
b)  You  are  interested  in  the  study  habits  of  grade  12  students,  so  you  interview  students  from  your  class.    
 
SAMPLING  BIAS  –  the  students  in  this  one  particular  class  may  not  represent  all  grade  12’s.  Students  with  
similar  traits  tend  to  take  the  same  types  of  classes.  This  is  a  convenience  sample  which  will  always  lead  to  
sample  bias.    
 
c)  You  ask  students  about  their  recycling  habits  on  behalf  of  the  Greenteam,  the  school  environment  club.    
 
RESPONSE  BIAS  –  since  your  are  asking  on  behalf  of  the  Greenteam,  students  may  feel  pressured  to  give  
answers  that  they  know  the  Greenteam  would  like  to  hear  instead  of  giving  honest  answers.  Anything  in  the  
survey  method  that  causes  people  to  give  incorrect  answers  creates  a  response  bias.    
 
d)  You  take  a  random  sample  of  5  students  from  each  block  A  class  to  determine  their  attitudes  toward  
the  new  school  attendance  policy.    
 
HOUSEHOLD  BIAS  –  not  all  classes  are  the  same  size.  Classes  should  be  surveyed  proportionately,  not  
equally.  Smaller  classes  are  over  represented  in  this  scenario.    
 
 
 
7)  When  a  phone  questionnaire  is  conducted,  many  people  with  call  display  will  not  answer  their  phone.  
What  kind  of  bias  does  this  represent?  What  can  be  done  to  minimize  this  kind  of  bias?  
 
NON-­‐RESPONSE  BIAS  –  company  can  block  name  from  call  display  to  reduce  this  bias  
 
8)  Suppose  you  want  to  know  the  average  amount  of  money  spent  by  the  fans  attending  opening  day  for  
the  Toronto  Blue  Jays.  You  get  permission  from  the  team’s  management  to  conduct  a  survey  at  the  
stadium,  but  they  will  not  allow  you  to  bother  the  fans  in  the  club  seating  or  box  seats  (the  most  
expensive  seating).  Using  a  computer,  you  randomly  select  500  seats  from  the  rest  o  the  stadium.  During  
the  game,  you  ask  the  fans  in  those  seats  how  much  they  spent  that  day.  What  type  of  bias  is  present  in  
this  survey  method?  
 
SAMPLING  BIAS  –  because  you  are  sampling  only  from  the  lower  priced  ticket  holders,  this  will  likely  
produce  an  estimate  that  is  too  small  and  not  representative  of  the  entire  population.    
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2.5  Worksheet–  Experiment  Design  
MDM4U  
Jensen  
 
Refer  to  part  2  of  the  lesson  for  help  with  the  following  questions  
 
1)  An  educator  wants  to  compare  the  effectiveness  of  computer  software  for  teaching  biology  with  that  of  
a  textbook  presentation.  She  gives  a  biology  pretest  to  each  of  a  group  of  high  school  juniors,  then  
randomly  divides  them  into  two  groups.  One  group  uses  the  computer,  and  the  other  studies  the  text.  At  
the  end  of  the  year,  she  tests  all  the  students  again  and  compares  the  increase  in  biology  test  scores  in  the  
two  groups.  Is  this  an  observational  study  or  an  experiment?  Justify  your  answer.  
 
Experiment,  because  students  were  randomly  assigned  to  the  different  teaching  methods.    
 
 
2)  One  study  of  cell  phones  and  the  risk  of  brain  cancer  looked  at  a  group  of  469  people  who  have  brain  
cancer.  The  investigators  matched  each  cancer  patient  with  a  person  of  the  same  age,  gender,  and  race  
who  did  not  have  brain  cancer,  then  asked  about  the  use  of  cell  phones.  The  results  suggested  that  the  use  
of  cell  phones  is  not  associated  with  risk  of  brain  cancer.  Is  this  an  observational  study  or  an  experiment?  
Justify  your  answer.    
 
Observational  study,  because  the  researchers  did  not  assign  people  to  either  use  or  not  use  of  cell  phones.    
 
3)  Do  smaller  classes  in  elementary  school  really  benefit  students  in  areas  such  as  scores  on  standardized  
tests,  staying  in  school,  and  going  on  to  college?  We  might  do  an  observational  study  that  compares  
students  who  happened  to  be  in  smaller  and  larger  classes  in  their  early  school  years.  Identify  a  potential  
variable  that  may  be  confounding  with  the  effects  of  small  classes.  
 
Type  of  school  and  socioeconomic  status  are  possible  confounding  variables.  Private  schools  tend  to  have  
smaller  class  sizes  and  students  that  come  from  families  with  higher  socioeconomic  status.  If  these  students  
do  better  in  the  future,  we  wouldn’t  know  if  it  was  due  to  smaller  class  sizes  or  type  of  school  or  
socioeconomic  status.    
 
4)  Ability  to  grow  in  shade  may  help  pines  found  in  the  dry  forests  of  Arizona  to  resist  drought.  How  well  
do  these  pines  grow  in  shade?  Investigators  planted  pine  seedlings  in  a  greenhouse  in  either  full  light,  
light  reduced  to  25%  of  normal  by  shade  cloth,  or  light  reduced  to  5%  of  normal.  At  the  end  of  the  study,  
they  dried  the  young  trees  and  weighed  them.    
 
a) Identify  the  experimental  units.  
 
Pine  seedlings  
 
b) What  are  the  explanatory  and  response  variables?  
 
Explanatory  variable:  light  intensity     Response  variable:  weight  of  tree  
 
c) What  are  the  treatments  used?  
 
Full  light,  25%  light,  and  5%  light  
 
 
5)  You  can  use  Skype  to  make  long-­‐distance  calls  over  the  Internet.  How  will  the  appearance  of  ads  
during  calls  affect  the  use  of  this  service?  Researchers  design  an  experiment  to  find  out.  They  recruit  300  
people  who  have  not  used  Skype  before  to  participate.  Some  people  get  the  current  version  of  Skype  with  
no  ads.  Others  see  ads  whenever  they  make  calls.  The  researchers  are  interested  in  frequency  and  length  
of  phone  calls.    
 
a) Identify  the  experimental  units.  
 
300  people  who  haven’t  used  Skype  before  
 
b) What  are  the  explanatory  and  response  variables?  
 
Explanatory  variable:  whether  ads  are  present  or  not      
Response  variable:  length  and  frequency  of  calls  
 
c) What  are  the  treatments  used?  
 
No  ads  shown  during  calls  and  ads  shown  during  calls  
 
 
Refer  to  part  3  of  the  lesson  for  help  with  the  following  questions  
 
6)  Dr.  Linda  Stern  and  her  colleagues  recruited  132  obese  adults  at  the  Philadelphia  Veterans  Affairs  
Medical  Center  in  Pennsylvania.  Half  the  participants  were  randomly  assigned  to  a  low-­‐carbohydrate  diet  
and  the  other  half  to  a  low-­‐fat  diet.  Researchers  measured  each  participant’s  change  in  weight  and  
cholesterol  level  after  six  months  and  again  after  one  year.  Explain  how  each  of  the  four  principles  of  
experimental  design  was  used  in  this  study.    
 
Comparison:  researchers  used  a  design  that  compares  low-­‐carb  diets  with  low-­‐fat  diets.    
Random  Assignment:  Subjects  were  randomly  assigned  to  one  of  the  two  diets.  
Control:  The  experiment  used  subjects  who  were  all  obese  at  the  beginning  of  the  study  and  who  all  lived  in  
the  same  area.  
Replication:  There  were  66  subjects  in  each  treatment  group  
 
 
7)  Does  day  care  help  low-­‐in-­‐come  children  stay  in  school  and  hold  good  jobs  later  in  life?  Carolina  
Abecedarian  Project  has  followed  a  group  of  111  children  since  1972.  Back  then,  these  individuals  were  
all  healthy  but  low-­‐income  infants  in  Chapel  Hill,  North  Carolina.  All  the  infants  received  nutritional  
supplements  and  help  from  social  workers.  Half  were  also  assigned  at  random  to  an  intensive  preschool  
program.  Explain  how  each  of  the  four  principles  of  experimental  design  was  used  in  this  study.    
 
Comparison:  Researchers  used  a  design  that  compared  children  who  were  assigned  to  an  intensive  pre-­‐
school  program  to  children  who  were  not  enrolled  in  an  intensive  preschool  program.    
Random  Assignment:  Subjects  were  randomly  assigned  to  be  enrolled  in  the  intensive  program  or  not.  
Control:  All  subjects  were  healthy,  low-­‐income,  and  from  the  same  area.  Also,  all  subjects  received  
nutritional  supplements  and  help  from  social  workers.    
Replication:  Over  50  subjects  in  each  group.    
 
 
 
 
8)  Researchers  in  Japan  conducted  an  experiment  on  13  individuals  who  were  extremely  allergic  to  
poison  ivy.  On  one  arm,  each  subject  was  rubbed  with  a  poison  ivy  leaf  and  told  the  leaf  was  harmless.  On  
the  other  arm,  each  subject  was  rubbed  with  a  harmless  leaf  and  told  it  was  poison  ivy.  All  the  subjects  
developed  a  rash  on  the  arm  where  the  harmless  leaf  was  rubbed.  Of  the  13  subjects,  11  did  not  have  any  
reaction  to  the  real  poison  ivy  leaf.  Explain  how  the  results  of  this  study  support  the  idea  of  a  placebo  
effect.    
 
The  subjects  developed  rashes  on  the  arm  exposed  to  the  placebo  (a  harmless  leaf)  simply  because  they  
thought  they  were  being  exposed  to  a  poison  ivy  leaf.  Likewise,  most  of  the  subjects  didn’t  develop  rashes  on  
the  arm  that  was  exposed  to  poison  ivy  because  they  didn’t  believe  they  were  being  exposed  to  the  real  thing.    
 
9)  The  progress  of  a  type  of  cancer  differs  in  women  and  men.  Researchers  want  to  design  an  experiment  
to  compare  tree  therapies  for  this  cancer.  They  recruit  500  male  and  300  female  patients  who  are  willing  
to  serve  as  subjects.  Which  are  the  block  in  this  experiment:  the  cancer  therapies  or  the  two  genders?  
Why?  
 
The  genders,  because  researchers  will  randomly  assign  all  three  therapies  within  each  gender.    
 
 
10)  A  nutrition  experimenter  intends  to  compare  the  weight  gain  of  newly  weaned  male  rats  fed  Diet  A  
with  that  of  rats  fed  Diet  B.  To  do  this,  she  will  feed  each  diet  to  10  rats.  She  has  available  10  rats  from  
one  litter  and  10  rats  from  a  second  litter.  Rats  in  the  first  litter  appear  to  be  slightly  healthier.    
 
a)  Why  would  it  be  poor  design  to  have  the  10  rats  from  Litter  1  be  fed  Diet  A,  and  the  10  rats  from  Litter  
2  be  fed  Diet  B?  
 
If  one  of  the  groups  gained  more  weight,  we  would  not  know  if  this  was  because  of  the  diet  or  because  of  
genetics  and  initial  health.  Genetics  and  diet  would  be  confounded.    
 
b)  Describe  a  better  design  for  this  experiment  
 
Use  a  randomized  block  design  with  the  litters  as  blocks.  For  each  of  the  litters,  randomly  assign  half  of  the  
rats  to  receive  Diet  A  and  the  other  half  to  receive  Diet  B.  This  will  allow  researchers  to  account  for  
differences  in  weight  gain  caused  by  differences  in  genetics.