FLDPSY

Quantitative Data Collection

• The HAU Psychology Society reviewers ensure consistency and quality, basing content only on instructor-given modules.

Five Steps in the Process of Data Collection

• Determine the participants to study.
• Obtain permissions necessary from relevant people and organizations.
• Consider the types of information to collect, ensuring it's available for quantitative research.
• Locate and select instruments for useful data.
• Administer the data collection process to collect data.

Identifying Participants

• Identify the unit of analysis at the individual, family, school, or district level.
• Specify the population and sample.

Populations and Samples

• Population: a group with a distinguishing characteristic.
• Target Population/Sampling Frame: the list of sampling units from which the sample is selected.
• Sample: participants selected from the target population used for generalization.

Probability Sampling

• Uses random selection to choose sample from the population.
• Aims to provide representative elements of populations.

Simple Random Sampling

• Each population element has an equal and independent chance of being chosen.

Methods to Select a Simple Random Sample

• Fishbowl Method: Assign numbers, draw from a container.
• Table of Random Numbers: Use a list of random numbers, selecting elements based on assigned numbers.
• Calculator: Use a calculator to generate random numbers.

Fishbowl Technique (Lottery Method)

• Write participant numbers on same-size papers.
• Put papers in a container for free movement.
• Shake thoroughly and draw a number.

Procedures in Fishbowl Technique

• Random Sample with Replacement: Return the drawn paper to the container.
• Random Sample without Replacement: Do not return the paper to the container, once drawn.

Table of Random Numbers Method

• Includes pre-generated numbers with no order or sequencing.
• Use: NRan# (ex. 1000Ran#)

Steps in the Table of Random Numbers Method

• Assign a number to each element of the accessible population.
• Enter the table at a random starting point.
• Systematically select numbers (up, down, left, right, or diagonally).
• Continue until the desired sample size is reached, circling the selected numbers.
• Skip any number encountered more than once.

Using a Calculator to Generate Random Numbers

• Use the function NRan# which, for example with 1000Ran#, will generate a random number.
  • N represents the population size.
  • Ran# is the calculator key used for the function.

Systematic Random Sampling

• Selects every kth element such as every fifth, eighth, or twenty-first element from the population.
• K = is the interval between elements
• N = is the population size
• n = is the sample size
• Formula to determine interval: K = N/n

Steps in Systematic Random Sampling

• Assign a number to each participant.
• Compute the interval K, using the formula above.
• Enumerate the elements and select a number for a random start.
• Select a number using a Table of Random Numbers for random start.

Multistage Cluster Sampling

• Also called Cluster Random Sampling.
• Large groups ("clusters") become the sampling units, eg Region, Province, Town and Barangay.

Using Power Analysis to Determine Sample Size

Power Analysis

• Power is the probability that the test correctly rejects a false null hypothesis.
• Power indicates if the test identifies a treatment effect if one exists.

Factors that Influence Power

• Effect Size: As the effect size increases so too does the probability of rejecting the null hypothesis.
• Sample Size: A larger sample produces greater power for a hypothesis test, used to achieve a reasonable probability for succesful research.
• Alpha Level: Reducing the alpha level reduces the power of the test.
• One vs. Two-Tailed Tests: Converts with a predicted direction, changing from two-tailed test to a one-tailed test increases power of tests.

Power Analysis using G*Power

• G*Power computes statistical power analyses for different t, F, x2, and z tests.
• G*Power computes effect sizes and to display graphically the results of power analyses.
• Statistical power analysis helps researchers determine the smallest sample size to detect an effect.

How to Determine Sample Size using G*Power

• Test Family: Exact
• Statistical Test: Correlation: Bivariate normal model
• Type of power analysis: A priori: Compute required sample size
• Tails: Two
• Correlation p H1: .25
• α err prob: .05 Power (1 - β err prob): .8
• Correlation p H0: 0
• An estimate of the size of correlation (r)
• A two tailed alpha value (a): This value is called the Type I error rate
• A beta value (β): This value is called the Type II error rate

Nonprobability Sampling

• Sample elements are chosen non-randomly.

Convenience Sampling

• Involves choosing readily available people or objects; also known as accidental or incidental sampling.

Snowball Sampling

• Participants identify others to become members of the sample; it is a type of network or chain sampling.

Quota Sampling

• Similar to stratified random sampling, but obtaining sample members is not random.

Purposive Sampling

• Involves “handpicking” of subjects; also known as judgmental sampling.

Considerations in Sample Size in Quantitative Studies

• Larger samples tend to be more representative.
• Larger samples decrease sampling error.
• Increased sample size diminishes the probability of a deviant sample.
• Power analysis determines adequate sample size.

Sampling Error

• Difference between random sample data and entire population data.

Sampling Bias

• Bias occurs when samples are not carefully selected.

Sample Size Estimates (Creswell, 2019)

• Approximately 15 participants per group in an experiment.
• Approximately 30 participants for a correlational study.
• Approximately 350 individuals for a survey study.

Sampling Designs in Qualitative Studies

Convenience and Snowball Sampling

• Volunteer Sampling: Researchers need participants to come forward.
• Nominated Sampling: Relies on nominations or referrals from existing participants.

Purposive Sampling

• Qualitative researchers purposefully select sample members based on emerging information needs.

Strategies for Purposive Sampling (Patton, 2015)

• Maximum variation sampling: selects cases with a range of variation on dimensions of interest.
• Homogenous sampling: deliberately reduces variation for focused inquiry.
• Typical case sampling: selects participants that illustrate what is typical or average.
• Extreme/deviant case sampling: learns from unusual and extreme informants (e.g., outstanding successes).
• Criterion sampling: studies cases meeting a predetermined criterion of importance.
• Sampling of confirming and disconfirming cases: --Confirming cases: fit researchers' ideas and offer credibility. --Disconfirming cases: challenge researchers' interpretations. -Theoretical Sampling
  • Used in grounded theory studies
  • Involves joint data collection, coding, and analysis, deciding what to collect next
  • Aims to discover categories, properties, and relationships in the substantive theory.

Considerations in Sample Size in Qualitative Studies

• Data saturation is the key concept, so final sampling decisions usually occur during data collection.
• Small number of informants can create a large sample because the incidents or experiences are the basis for analysis.
• Ethnographic studies: 25-50 informants.
• Phenomenology: 10 or lower.
• Grounded theory studies: 20-30 people.

Sample Description in Qualitative Studies

• Reports should describe the sampling procedure, population, selection criteria and rationale for sample size.

Obtaining Approval

• Required from Institutions, Specific Research sites, research participants, Parents of participants and the University or college institutional review board

Review Board Approval

• An IRB is a committee made up of faculty members

Obtaining Approval from Review Boards

• Start by learning about the board's review process on your campus.
• Determine what information is required by the review board about your project.
• Create an informed consent form for participants to sign.

Determine Type of Information to Collect

The Flow of Activities in Collecting Data

• Identify the variable
• Operationally define the variable
• Locate data (measures, observations, and documents with questions and scales) Types of Data

Scales

Individual Performance

• Types of Tests, Instruments, or Documents to Collect Data -- Achievement test: norm-referenced tests -- Criterion-referenced tests -- Intelligence test

Individual Attitude

• Types of Tests, Instruments, or Documents to Collect Data -- Affective Scale

Of Individual Behavior

• Types of Tests, Instruments, or Documents to Collect Data --Behavioral Checklist

Instruments for Collecting Data

• Types of Data Facts, data collection or data records

Instruments for Collecting Data

• Types of Data

Instruments for Collecting Data

• aptitude tests

Selecting an Instrument

Steps in Developing or Constructing an Instrument

• Phase 1: Begin by stating the purpose of the test and defining the target group
• Phase 2: Construct the data collecting method while content validating and evaluating
• Phase 3: Implement a quantitative evaluation
• Phase 4: Proceed to testing as a method for continuous validation

Criteria for Choosing a Good Instrument

• New, updated, and recent material
• If reviews exist, it means that other have sought to seriously question reliability or validity.
• Is this a well sited and well known work for information about usage?
• Review for available information about reliability

Modifying an instrument

• means locating an existing instrument, obtaining permission to change it, and making changes in it to fit your requirements.

Reliability and its variables

• A variable that affects reliability will result with unstable or inconsistent data.
• Scores on instruments must be precise for scores to provide correct feedback.

Types of Reliability

• Test-retest reliability : One version, two times at different intervals
• Alternate forms reliability : Two versions, each only administered once
• Alternate forms and test retest reliability: Two versions, twice at intervals.
• Interrater reliability - Observed by other raters.
• Internal consistency reliability - Once completed and consistent throughout.

Validity

• demonstrates the test interpretation.

Types of Scales Used in Quantitative Research

• Nominal scale (uses categories)
• Ordinal scale (uses categories that imply or express rank order)
• Quasi-interval or interval/ratio scale (uses continuous equal intervals)

Survey Research- Three aspects of using mailed questionnaires

• Cover letter
• Construction
• Analyzed using statistical procedures

Parts of a Mailed Questionnaire

• A cover letter.
• Closed ended questions, background information.
• Closed ended practices asking about behaviors.
• Open ended, allowing the respondent's unique insights.
• Thanks to the interview by participant.

Common Problems in Item Construction

Problem Example Solution Examples

1. Unclear : Be direct, and don’t hesitate to be obvious.
2. Multiple Questions
3. Length: Lengthy descriptions may cause confusion.
4. Negation/Sarcasm : Don’t be sarcastic or rude. The respondents may not be honest if they feel “forced.”

Variables in The Collection Process

Types of scales

1 :Categorical scales have two types: nominal and ordinal scales. 2 :Continuous scales (often called scale scores or rating scales) have two types: interval and ratio scales. Researchers use nominal scales (or categorical scales) to provide response options where participants check one or more categories that describe their traits, attributes, or