Correlational Research
Research Methodology – ENGL 6001




by Ihsan Ibadurrahman – G1025429
What is a correlational
research?
   It aims to look at an empirical
    relationship between two variables such
    that (1) changes in one are associated
    with changes in the other or (2) particular
    attributes of one variable are associated
    with particular attributes of the other.
    (Babbie, 2007)

   Not to be confused with „association‟ –
    Correlation has a specific technical
    meaning and needs statistical
    requirements for it, whereas association
    is a more general idea. (Neuman, 2011)
Why do we use a correlational
research?
 As a first step prior to experimentation
 As one of the criteria used to
  determine Nomothetic Casuality.
 When experiments cannot be
  conducted (for ethical or practical
  reasons)
A word of caution
   Correlational studies can suggest that
    there is a relationship between two
    variables, but they cannot prove that
    one variable causes a change in
    another variable. In other words,
    correlation does not equal causation.


Out-of-class language       English performance in
       learning                      class
How to conduct a correlational
research?
 Variables are identified
 Questions and/or hypotheses are
  stated
 A sample is selected
 Data are collected
 Correlations are calculated
 Results are reported
Step 1: Identifying Variables
   “Predictor” variable – the variable(s)
    that are believed to predict the
    outcome, also called independent
    variable

   “Criterion” variable – the variable to be
    predicted, the outcome, also called the
    dependent variable.
Step 2: Stating Questions
   Is level of education (independent
    variable) related to family income
    (dependent variable)?

   Do people who eat more eggs
    (independent variable) have higher
    cholesterol levels (dependent variable)?

   Do students who employ out-of-class
    strategies (independent variable) more
    often have higher English scores
    (dependent variable)?
Step 3: Sampling
 Random Sampling
 Convenient / Purposeful sampling
 A minimum of 30 samples is required
Step 4: Gathering data
   Naturalistic Observation
    Naturalistic observation involves observing and
    recording the variables of interest in the natural
    environment without interference or manipulation by the
    experimenter.

   The Survey Method
    In this method, a random sample of participants
    completes a survey, test, or questionnaire that relates to
    the variables of interest.

   Archival Research
    Archival research is performed by analyzing studies
    conducted by other researchers or by looking at
    historical patient records.
Step 5: Calculating
correlations
   To calculate a numerical value of a
    correlation we can use Pearson’s
    product moment correlation
    cooficcient or correlation coefficient
    with the symbol of the lowercase letter
    „r‟.

   A correlation coofficient ranges from -
    1.0 to +1.0, with -1.0 indicating a
    perfect linear negative correlation and
    +1.0 a perfect linear positive
    correlation.
Interpretation of the Strength of
Correlations

 00 - .20 – Very Weak
 .21 - .40 – Weak
 .41 - .60 – Moderate
 .61 - .80 – Strong
 .81 – 1.00 - Very Strong

    Different statisticians may
    have similar but slightly
    different scales.
Step 6: Reporting results
   Positive Correlations: Both variables increase
    or decrease at the same time. A correlation
    coefficient close to +1.00 indicates a strong
    positive correlation.

   Negative Correlations: Indicates that as the
    amount of one variable increases, the other
    decreases (and vice versa). A correlation
    coefficient close to -1.00 indicates a strong
    negative correlation.

   No Correlation: Indicates no relationship
    between the two variables. A correlation
    coefficient of 0 indicates no correlation.
Correlations
      Scatter plots are often used to
       depict correlations
                       6000

                       5000
    Calories per day




                                                                 This chart shows
                       4000
                                                                 a strong positive
                       3000                                      correlation

                       2000

                       1000
                         0
                          100   150   200    250     300   350       400
                                            Weight
Correlations
        Scatter plots are often used to
         depict correlations
                                  160                                This chart shows
    Minutes of Exercise per day




                                  140                                a strong
                                                                     negative
                                  120
                                                                     correlation
                                  100
                                  80
                                  60
                                  40
                                  20
                                   0
                                    100   150   200    250     300   350    400
                                                      Weight
Correlations
       Scatter plots are often used to
        depict correlations
                              45
    Miles from Krispy Creme




                              40
                              35
                                                                       This chart shows
                              30
                                                                       virtually no
                              25                                       correlation
                              20
                              15
                              10
                               5
                               0
                                100   150   200    250     300   350      400
                                                  Weight
How to calculate correlations
   Excel has a statistical function. It calculates
    Pearson Product Moment correlations.

   SPSS (a statistical software program for
    personal computers used by graduate
    students) calculates correlations.

Correlational research

  • 1.
    Correlational Research Research Methodology– ENGL 6001 by Ihsan Ibadurrahman – G1025429
  • 2.
    What is acorrelational research?  It aims to look at an empirical relationship between two variables such that (1) changes in one are associated with changes in the other or (2) particular attributes of one variable are associated with particular attributes of the other. (Babbie, 2007)  Not to be confused with „association‟ – Correlation has a specific technical meaning and needs statistical requirements for it, whereas association is a more general idea. (Neuman, 2011)
  • 3.
    Why do weuse a correlational research?  As a first step prior to experimentation  As one of the criteria used to determine Nomothetic Casuality.  When experiments cannot be conducted (for ethical or practical reasons)
  • 4.
    A word ofcaution  Correlational studies can suggest that there is a relationship between two variables, but they cannot prove that one variable causes a change in another variable. In other words, correlation does not equal causation. Out-of-class language English performance in learning class
  • 5.
    How to conducta correlational research?  Variables are identified  Questions and/or hypotheses are stated  A sample is selected  Data are collected  Correlations are calculated  Results are reported
  • 6.
    Step 1: IdentifyingVariables  “Predictor” variable – the variable(s) that are believed to predict the outcome, also called independent variable  “Criterion” variable – the variable to be predicted, the outcome, also called the dependent variable.
  • 7.
    Step 2: StatingQuestions  Is level of education (independent variable) related to family income (dependent variable)?  Do people who eat more eggs (independent variable) have higher cholesterol levels (dependent variable)?  Do students who employ out-of-class strategies (independent variable) more often have higher English scores (dependent variable)?
  • 8.
    Step 3: Sampling Random Sampling  Convenient / Purposeful sampling  A minimum of 30 samples is required
  • 9.
    Step 4: Gatheringdata  Naturalistic Observation Naturalistic observation involves observing and recording the variables of interest in the natural environment without interference or manipulation by the experimenter.  The Survey Method In this method, a random sample of participants completes a survey, test, or questionnaire that relates to the variables of interest.  Archival Research Archival research is performed by analyzing studies conducted by other researchers or by looking at historical patient records.
  • 10.
    Step 5: Calculating correlations  To calculate a numerical value of a correlation we can use Pearson’s product moment correlation cooficcient or correlation coefficient with the symbol of the lowercase letter „r‟.  A correlation coofficient ranges from - 1.0 to +1.0, with -1.0 indicating a perfect linear negative correlation and +1.0 a perfect linear positive correlation.
  • 11.
    Interpretation of theStrength of Correlations  00 - .20 – Very Weak  .21 - .40 – Weak  .41 - .60 – Moderate  .61 - .80 – Strong  .81 – 1.00 - Very Strong Different statisticians may have similar but slightly different scales.
  • 12.
    Step 6: Reportingresults  Positive Correlations: Both variables increase or decrease at the same time. A correlation coefficient close to +1.00 indicates a strong positive correlation.  Negative Correlations: Indicates that as the amount of one variable increases, the other decreases (and vice versa). A correlation coefficient close to -1.00 indicates a strong negative correlation.  No Correlation: Indicates no relationship between the two variables. A correlation coefficient of 0 indicates no correlation.
  • 13.
    Correlations  Scatter plots are often used to depict correlations 6000 5000 Calories per day This chart shows 4000 a strong positive 3000 correlation 2000 1000 0 100 150 200 250 300 350 400 Weight
  • 14.
    Correlations  Scatter plots are often used to depict correlations 160 This chart shows Minutes of Exercise per day 140 a strong negative 120 correlation 100 80 60 40 20 0 100 150 200 250 300 350 400 Weight
  • 15.
    Correlations  Scatter plots are often used to depict correlations 45 Miles from Krispy Creme 40 35 This chart shows 30 virtually no 25 correlation 20 15 10 5 0 100 150 200 250 300 350 400 Weight
  • 16.
    How to calculatecorrelations  Excel has a statistical function. It calculates Pearson Product Moment correlations.  SPSS (a statistical software program for personal computers used by graduate students) calculates correlations.