Fundamentals of Statistics: Concepts and Applications

Nature and Scope of Statistics

Definition: The science of collecting, organizing, analyzing, and interpreting numerical data to understand behavior.

Key Points

• Aggregate of Facts: It deals with groups (populations), not single individuals.
• Variability: It exists because people are different; if everyone were the same, we wouldn't need it.
• Art & Science: It uses mathematical rules (Science) but requires judgment to choose the right test (Art).
• Scope: Used in clinical psychology (testing treatments), industrial psychology (hiring), and research.

Descriptive vs. Inferential Statistics

Descriptive Statistics

Summarizes the data you have in front of you.

• Tools: Mean, Standard Deviation (SD), Graphs.
• Example: Finding the average age of 50 students in your room.

Inferential Statistics

Makes predictions about a large population based on a small sample.

• Tools: T-tests, Correlation, ANOVA.
• Example: Testing a new therapy on 20 people to see if it will work for all depressed patients.

Scales of Measurement (NOIR)

• Nominal: Categories only (e.g., Blood group, Gender). No math possible.
• Ordinal: Ranking (e.g., 1st, 2nd, 3rd place). Distance between ranks is unknown.
• Interval: Equal distance between points, no true zero (e.g., Temperature, IQ).
• Ratio: Equal distance and true zero (e.g., Height, Weight, Time).

Graphical Representation

• Histogram: Bars touch; used for continuous class intervals.
• Frequency Polygon: Line graph connecting midpoints of intervals; shows the shape of data.
• Bar Graph: Bars do not touch; used for discrete categories (e.g., Male vs. Female).

Merits & Limitations of Standard Deviation

• Merits: Most stable measure of variability; uses every score in the data; essential for further tests like Z-scores.
• Limitations: Difficult to calculate manually; heavily influenced by extreme outlier scores.

Divergence from Normality

Normal Curve (NPC)

Bell-shaped, symmetrical, Mean = Median = Mode.

Skewness (Horizontal Distortion)

• Positive: Tail to the right (scores are low).
• Negative: Tail to the left (scores are high).

Kurtosis (Vertical Distortion)

• Leptokurtic: Very peaked/thin.
• Platykurtic: Flat/broad.

The Normal Probability Curve (NPC)

Definition: A symmetrical, bell-shaped theoretical distribution where the Mean, Median, and Mode all coincide at the center.

Key Properties

• Symmetry: The left half is a mirror image of the right half.
• Asymptotic: The ends of the curve approach the X-axis but never actually touch it.
• Total Area: The area under the curve is always 1.00 or 100%.
• Standard Deviation Limits: 68.26% of cases fall between ±1 SD; 95.44% between ±2 SD.

Percentiles and Percentile Ranks

Percentile (P): A point on the score scale below which a given percentage of cases fall.

• Example: If the 70th percentile score is 85, it means 70% of people scored below 85.

Percentile Rank (PR): The percentage of cases in a distribution that fall below a specific score.

Key Differences

• A Percentile is a score (e.g., "His IQ score is 115").
• A Percentile Rank is a percentage (e.g., "He is in the 90th percentile").

Calculation Note: To find these, you MUST use Cumulative Frequency (cf).