POLSC 193 Midterms

Save Me

What are the parts of a research proposal?

1. introduction

• problem statement / motivation

• research questions

• significance and objectives

• scope and limitations
  

2. review of related literature and research framework

3. methodology

Survey Research

a method data collection in which information is obtained directly from individual persons who are selected so as to provide a basis for making inferences about some larger population. used to ask respondents about their respective behaviors, attitudes, beliefs, opinions, characteristics, expectations, self-classification, and knowledge

Descriptive and Inferential

two types of statistics

What are the Fundamental Ethical Principles?

• respect for persons

• beneficence

• justice

How do we protect our research participants?

using an Informed Consent Form

What are the levels of measurement?

Qualitative/Categorical

• Nominal: numbers or symbols are assigned to a set of categories for the purpose of naming, labeling, or classifying observations; cannot be ranked-ordered

• Ordinal: nominal levels that can be ranked from low to high; does NOT include the magnitude of differences between numbers

Quantitative

• Interval-Ratio: all cases are expressed in the same units

  • INTERVAL: equal intervals = equal differences

  • RATIO: like interval, but ratios if scores must also make sense

Measurement Error

• Differences in the values assigned to cases that are attributable to anything other than real differences; random vs systematic errors

Reliability

assessment of how stable the values the measure yields are; suggests that repeated, stable outcomes are the same under identical of similar conditions

Improving Reliability

clearly conceptualize all constructs; increase the level of measurement; use multiple indicators of a value; use pre-tests, pilot studies, and replication

Validity

Refers to how well the conceptual and operational definitions mesh with each other: the better the fit, the greater the measurement validity; the extent to which our measures correspond to the concepts they are intended to reflect

Types of Validity

Face Validity

• A judgment by the scientific community that the indicator really measures the construct

Content Validity

• Is the full content of a definition represented in a measure?

Construct Validity

• Internal: infer validity of the indicator from its relationship to other indicators of the same concept using multiple indicators

• External: infer validity of the indicator from its relationship to indicators of other concepts to which the concept being measured should theoretically be related

Survey Research

• the purpose is to discover relative incidence, distribution, and interrelations of sociological and psychological variables (Kerlinger 1986)

• defined as “the collection of information from a sample of individuals through their responses to questions” (Check & Schutt, 2012)

• method of data collection in which information is obtained directly from individual persons who are selected so as to provide a basis for making inferences about some larger population

• often in the form of a structured questionnaire

Surveys can be used to/for…

• Describe. Surveys can be used to describe and summarize phenomena. The goal is to get a precise measurement of such phenomena.

• Causal Explanations. Surveys measure associations between variables.

  • in statistics, we are looking for CORRELATION, not causation

• Evaluation. Surveys can be useful for determining the degree to which a desired objective is attained as a result of a planned intervention.

• Prediction. Survey data can be used to forecast future events.

Survey Research steps?

1. purpose & objective

2. target population

3. data gathering techniques

4. sampling

5. data gathering proper

6. data cleaning and processing

7. data analysis

8. report writing

Advantages of a survey research design?

• inferences are not based on theory or dogma but it is based on facts

• ensures greater objectivity and reliability

• important aspect is its versatility

• cost-effective

Weaknesses of a survey research design?

• its reliability and validity is based on the honesty and efficiency of the researcher

• mostly based on samples, so there is always a possibility of sampling error

• as data is collected from primary sources, the feasibility depends upon the willingness and cooperation of the respondent

• there is a possibility of response error, due to respondents’ untrue/misleading answers

Quantitative Research Questions

• inquire about the relationships among variables that the investigator seeks to know

• hypotheses are predictions the researcher makes about the expected outcomes of relationships among variables

• follows from a test of theory, and the specifications of research questions or hypotheses follow from the relationship among variables in the theory

• use the same pattern of word order in the questions or hypotheses to enable a reader to easily identify the major variables

• independent and dependent variables must be measured separately, and not measured on the same concept (to ensure falsifiability of hypotheses)

What type of research asks the question: How are the variables distributed?

Descriptive

What type of research asks the question: How are the variables related?

Descriptive-Explanatory

What type of research asks the question: Why are the variables distributed and related this way?

Explanatory

Null Hypothesis (H0)

• a statement of “no difference” or “no relationship” that contradicts the alternative hypothesis and is always expressed in terms of population parameters

• there are no differences between two groups or there is no observed differences between the two populations

• there is no correlation/relationship between two variables

Conceptual Definition

• a careful, systematic definition of a construct that is explicitly written to clarify one’s thinking

• often linked to other concepts or theoretical statements (Neuman, 2014)

Operational Definition

• definition of a variable in terms of the specific activities to measure or indicate it with empirical evidence (Neuman, 2014)

Frequency Distributions

• a summary of the responses to the categories of a variable

• can be used for all levels of measurements, with reservations

Examples of missing variables?

the participant…

• skipped the question

• didn’t know what to put

• put down NAP, NA/NI, DK

What are the measures of central tendency?

• mean

• median

• mode

Mode

• most frequently occurring value in a distribution

• abbreviated as mo

• only measure of central tendency appropriate for nominal-level variables

Bimodal

sometimes there is more than one mode

Formula for position of the Median?

(N+1)/2

Mean

• the average score

• only appropriate for interval and ratio level variables

How do you find the mean?

divide the sum of the raw scores in a set of scores with the total number of scores in a set

Range

• the difference between the highest and lowest scores in a distribution

How do you find the range?

R = H - L

• R = range

• H = highest score in a distribution

• L = lowest score in a distribution

Grouped Frequency Distribution

• condensing the separate scores into a number of smaller categories or groups, each containing more than one score value

Class Interval

• a way of categorizing data (in 5s, 10s, etc.); preferably a whole number

Class Limits

• located at the point halfway between adjacent class intervals and serves to close the gap between them

• formula: (HL - LL)/K

  • K = number of classes

  • HL = higher limit; LL = lower limit

Proportions and Percentages

• a way of standardizing the frequency distributions for size, when you want to compare the distribution of responses between groups

• formula: P = f/N

  • f = frequency

  • N = total number of cases

Variance

• squaring deviations results in positive calculations

• just… just look at the image man

Inter-quartile range

• anak ng butiki…

Inferential Statistics

• population = consists of a set of individuals who share at least one characteristic; the entire group that you want to draw conclusions about

• sample = a smaller number of individuals from the population; a specific group of individuals that you will collect data from

Two Types of Sampling

• probability sampling

  • involves the equal probability of selection method (or EPSEM)

  • allows us to make stronger and more reliable statistical inferences from sample to population

• non-probability sampling

  • involves non-random selection methods, which allows us to easily gather data

  • used when sampling frames are not available

  • making inferences and predictions is not possible

Types of Probability Sampling

• simple random sampling

• stratified random sampling

• multi-stage cluster random sampling

Types of Non-probability Sampling

• convenience sampling

• purposive sampling

• quota sampling

• snowball sampling

Sampling Formula

n = N/(1+Ne²)

• n = sample size

• N = total population size

• e = margin of error

Simple Random Sampling

• obtain a list of the population

• assign a unique identifying number to each name

• with eyes closed, “enter” the table of random numbers

• move in any direction and, depending on how you assigned a number to your list of population, take the names corresponding to the numbers selected

Stratified Random Sampling

• involves dividing the population into strata (homogenous subgroups)

• proportionate vs disproportionate

Multi-stage Cluster Random Sampling

• used to minimize costs of large surveys

• random sampling of a primarily unit (e.g. barangays, residential buildings, etc.) and then random sampling of members within the cluster

Convenience Sampling

• includes the individuals who happen to be the most accessible to the researcher

• or participants volunteer their participation

Purposive Sampling

• logic, common sense, or sound judgement can be used to select a sample

Quota Sampling

• like stratified sampling, individuals are divided into sub-groups and then judgement is used to select respondents based on a quota

Snowball Sampling

• referral system; drawing (future) samples based on network of existing respondents

• can be linear or discriminative snowball

Normal Distribution

• a type of probability distribution that is characterized by a bell-shaped curve

• an ideal model of how data is distributed in the population

• to what extent to data from our sample resemble the normal curve (i.e., the population)?

Z-score / standard score

• gives us the exact (if not exact, a good approximation) number of standard deviations a particular raw score is from the mean

• standardizes a score with respect to the other scores in the group

• formula: z = X - mean / standard deviation

Sampling Distribution of Means

• if we repeatedly get samples from our population and plot their means, the data that we will get also comes to approximate the normal curve

Statistical Hypothesis Testing

• allows us to evaluate hypotheses about population parameters based on sample statistics

• assumptions:

  • the sample was randomly selected

  • the variable is measured at the interval-ratio level

  • we can’t assume that the population is normally distributed

Alternative Hypothesis (H1)

• statement reflecting the substantive hypothesis

• specifies that the population is one of the following:

  • not equal to some specified value

  • greater than some specified value

  • less than some specified value

One-tailed Test

• a hypothesis test in which the alternative is stated in such a way that the probability of making a Type I error is entirely in one tail of a sampling distribution

> greater than some specified value
> less than some specified value

Right-tailed Test

• a one-tailed test in which the sample outcome is hypothesized to be at the right tail of the sampling distribution

> greater than some specified value

Two Types of Tests of Significance

• Parametric

• Non-parametric

Parametric tests of significance

• a normal distribution

• interval-ratio level of measurement for variables; and

• a large sample size

• examples: z-test, one sample/two samples t-test, ANOVA, and Pearson’s correlation

Non-parametric tests of significance

• distribution is skewed or normal distribution is not assumed;

• the variables are measured at the nominal or ordinal level; or when

• we do not have large sample

• examples: chi-square

Chi-square

• expected to refer to the null hypothesis

• a non-parametric test of significance that allows us to test the difference between f0 and fe

• a comparison of frequencies, not means

• the greater the difference between the f0 (observed frequency) and fe (expected frequency_, the more likely we can reject the null hypothesis

Chi-square example

anak ng putangina just look at the slide bro (more examples sa discord)