Nov 15- Statistical Evaluation of Data

Overview of Statistical Evaluation of Data

Four Levels of Measurement Scales:
- Nominal: Categorizes data without any quantitative value (e.g., gender, race).
- Ordinal: Arranges categories in a meaningful order but does not specify the distance between them (e.g., rankings).
- Interval: Numeric scales with equal intervals, but no true zero point (e.g., temperature in °C or °F).
- Ratio: Similar to interval with a true zero, allowing for the comparison of magnitudes (e.g., height, weight).

Definition: Methods to organize, summarize, and simplify research study outcomes.
Techniques:
- Tables or graphs to represent data visually.
- Numerical measures for data summarization.
Frequency Distribution:
- Illustrates how often each score appears within the data (e.g., exam scores).
- Helps identify patterns and distributions.

Pie Chart: Represents relative proportions; best for nominal data.
- Example: Gender distribution in a sample.
Bar Graph: Displays each distinct category with separate bars; suitable for nominal and ordinal data.
- Example: Responses across different survey questions.
Histogram: Shows frequency of continuous data using adjacent bars; for interval and ratio scales.
- Example: Exam grade distributions.
Polygon: Connects points representing frequencies across categories, suitable for interval and ratio scales.

Defines the center point of a dataset; includes:
- Mean: The average, calculated by adding scores and dividing by the number of scores.
- Median: The middle score when data is ordered.
- Mode: The most frequently occurring score.
Applications in determining average tendencies in data sets.

Standard Deviation: Indicates the average deviation of scores from the mean; provides insight into data dispersion.
- Example: Researchers analyze spread of ages in a sample to interpret variability in demographics.
Variance: The average of the squared deviations from the mean; assists in understanding data spread.
Steps to Calculate Standard Deviation:
1. Compute the mean.
2. Determine the deviation of each score from the mean.
3. Square each deviation.
4. Calculate the variance by averaging squared deviations.
5. Take the square root of the variance to get standard deviation.

Definition: Allows researchers to generalize findings from sample data to a larger population.
Importance in evaluating whether observed differences in sample means reflect true population differences.
Statistical Testing:
- Essential for estimating the likelihood that observed differences arose by chance.
- A significant result (e.g., p < .05) indicates a statistically significant difference between groups.
- Hypothesis Testing:
  - Null Hypothesis (H0): Assumes no effect or difference (e.g., no difference in treatment effects).
  - Alternative Hypothesis (Ha): Proposes a significant effect or difference (e.g., new medication reduces symptoms).
- Significance Levels: Threshold for testing, indicating the likelihood of error in rejecting H0 (e.g., .05 = 5% chance).

Sample Size: Larger samples generally yield more reliable results.
Variance: Lower variance increases the likelihood of finding statistically significant results.
- Comparison of research studies with different variances illustrates how variance can obscure mean differences.

Structure for reporting t-tests, including the depiction of means, standard deviations, t-values, and p-values.
Importance of clear reporting in the methods and results sections of research papers.
- Example reporting: "Participants who exercised (M = 4.47, SD = 0.63) had significantly higher scores than those who did not (M = 2.63, SD = 0.62, t(58) = 11.42, p < .001)."

SPSS Tabs:
- Variable View: Define the variables used in the analysis.
- Data View: Input and manage collected data.
Steps for conducting t-tests and reporting in SPSS focusing on entering variables, analyzing outputs, and generating descriptive statistics.