Comprehensive Statistics and Combinatorial Analysis

Flashcards covering introductory statistics, data types, data presentation, scales of measurement, measures of central tendency, dispersion, distribution shape (skewness and kurtosis), and combinatorial counting principles.

How is statistics defined, and into what two broad branches do statistical methods fall?

Statistics is defined as a branch of Mathematics (or a science of data) concerned with collecting, organizing, summarizing, presenting, analyzing, interpreting, and drawing conclusions or inferences from data. Its methods fall into two broad areas: descriptive statistics and inferential statistics.

What is the primary operational distinction between descriptive statistics and inferential statistics?

Descriptive statistics merely describes, organizes, or summarizes actual available data without drawing broader conclusions (e.g., mean blood pressure of a specific group of patients). Inferential statistics uses inductive reasoning to draw inferences and generalizations about a larger population based on observed sample data (e.g., hypothesis testing, ANOVA, and regression).

What are the core characteristics and common collection methods of qualitative data?

Qualitative data are non-numerical, descriptive, and expressed in words or sentences capturing feelings, emotions, or subjective perceptions through open-ended questions. Collection methods include focus groups, group discussions, and interactive interviews, which are documented using field notes, sketches, audiotapes, and photographs. Findings are indicative only of the participants and not generalizable to a broader population.

What are the core characteristics of quantitative data, and how are participants selected if population generalization is intended?

Quantitative data are numerical, mathematically computable, standardized, and gathered using structured instruments with closed-ended questions (e.g., experiments, clinical trials, and surveys). When the intent is to generalize findings from participants to a larger population, researchers employ probability sampling (random sampling).

What is a mixed methods approach, and what are three primary challenges associated with it?

A mixed methods approach combines both qualitative and quantitative research data, techniques, and methods within a single research framework to maximize strengths and reduce single-design weaknesses. Its three major challenges are: 1. Designing complementary qualitative and quantitative research questions; 2. Time-intensive data collection and analysis; and 3. Decisions regarding which research methods to combine.

What distinguishes primary data from secondary data in research?

Primary data are collected firsthand by the investigator specifically for the study under investigation, are unpublished, authentic, and unaltered by others. Secondary data are collected by a third party for another purpose and previously published (e.g., censuses, organizational records, and journal articles).

What four classifications are encountered in the process of data tabulation?

Data tabulation involves four types of classification: 1. Geographical (area-wise); 2. Chronological (on the basis of time); 3. Qualitative (according to attributes); and 4. Quantitative (according to magnitude).

What are the basic rules regarding the coordinate axes and point of intersection when constructing graphs?

The vertical axis is always labeled as the $y$-axis and called the 'ordinate', presenting frequencies. The horizontal axis is always labeled as the $x$-axis and called the 'abscissa', presenting values of variables. They meet at right angles at a point called the origin ($o$).

What type of correlation between two variables is illustrated in the scatter diagram shown below?

Zero or no correlation (there is no discernible relationship between the variables as the plotted points are randomly dispersed).

What is the distinction between positive correlation and negative correlation in a scatter diagram?

Positive correlation occurs when one variable increases as the other variable also increases (e.g., studying more leads to higher test scores). Negative correlation occurs when one variable increases while the other variable decreases, or vice versa (e.g., working more hours reduces free time).

How do bar charts and histograms differ in their data presentation and structure?

Bar charts display discrete categories or nominal data with distinct spaces between bars, with the vertical axis representing values such as group means. Histograms display continuous data with connected bars touching each other, where the horizontal axis represents class intervals and the vertical axis represents frequency.

What graphic device is shown below, and how is it constructed?

A frequency polygon. It is constructed from a histogram by replacing the vertical bars with straight line segments connecting the midpoints of the tops of the bars.

What is a pie chart and what do its components represent?

A pie chart is a circular graph that displays data divided into slices proportional to the fraction of the whole in each category. The entire pie represents $100\%$ of the whole, while each slice represents a proportional relative share of the total.

What are the four primary scales of measurement in statistics?

The four primary scales of measurement are the nominal scale, ordinal scale, interval scale, and ratio scale.

What are the defining characteristics and permissible statistical operations of the nominal scale?

The nominal scale categorizes data into distinct groups with no inherent order, ranking, or hierarchy, assigning numbers or labels purely for identification (e.g., gender, blood type, jersey numbers). The only permissible operations are counting frequencies and computing the mode; arithmetic operations cannot be performed.

What distinguishes the ordinal scale from the nominal scale?

The ordinal scale organizes data into categories that possess a meaningful order or rank (e.g., customer satisfaction ratings, race finishing positions), but the intervals between ranks are unequal and non-standardized. Permissible operations include rank comparisons ($>$ or $<$) and median calculation, but not addition or subtraction.

Why can ratios not be interpreted on an interval scale?

An interval scale has equal intervals between values but lacks a true zero point, meaning zero does not represent the complete absence of the attribute (e.g., $0^{\circ}\text{C}$ does not mean 'no temperature', and an IQ score of $0$ does not mean 'no intelligence'). Consequently, addition and subtraction are valid, but ratios, multiplication, and division are not meaningful.

What defines a ratio scale, and which mathematical operations are permissible on it?

A ratio scale possesses equal intervals between values and an absolute, meaningful zero point representing 'none' or total absence (e.g., height, weight, age, and income). It permits all mathematical operations, including addition, subtraction, multiplication, and division, making ratio comparisons completely valid.

What does central tendency refer to, and what are its three most important measures?

Central tendency refers to the location of a distribution. Its three most important measures are the mean, the median, and the mode.

What are the formulas for calculating the population mean and sample mean for ungrouped data?

For a population: $\mu = \frac{\sum X}{N}$. For a sample: $\bar{X} = \frac{\sum X}{n}$, where $\sum X$ represents the sum of all observations, and $N$ and $n$ denote the total number of observations in the population and sample, respectively.

What is the formula for calculating the arithmetic mean for grouped frequency data?

For a population: $\mu = \frac{\sum fX}{N}$. For a sample: $\bar{X} = \frac{\sum fX}{n}$, where $f$ represents the frequency of each class, $X$ represents the class midpoint, and $\sum fX$ is the sum of the frequency-midpoint products.

How is the median calculated for ungrouped data?

All items are first arranged in ascending or descending order. The median is located at the $\left(\frac{N + 1}{2}\right)\text{th}$ position in the array. If the number of items is even, the median is the arithmetic average of the two middle items.

What formula calculates the median for grouped data, and what does each term represent?

$\text{Median} = L + \frac{\frac{n}{2} - F}{f_m}c$, where $L$ is the lower limit of the median class, $n$ is the total number of observations, $F$ is the cumulative frequency up to but not including the median class, $f_m$ is the frequency of the median class, and $c$ is the class interval width.

What is the formula for calculating the mode of grouped data, and what do $d_1$ and $d_2$ represent?

$\text{Mode} = L + \frac{d_1}{d_1 + d_2}c$, where $L$ is the lower limit of the modal class, $c$ is the class interval width, $d_1$ is the frequency of the modal class minus the frequency of the preceding class, and $d_2$ is the frequency of the modal class minus the frequency of the following class.

How do extreme values differentially affect the mean, median, and mode?

The arithmetic mean is heavily affected by extreme values (outliers) in a data set, whereas the median and mode are resistant and not affected by extreme values.

Besides the arithmetic mean, median, and mode, what are three other measures of central tendency?

1. Weighted mean; 2. Geometric mean; and 3. Harmonic mean.

What are the formulas for average deviation (mean absolute deviation) for ungrouped population and sample data?

For a population: $AD = \frac{\sum |X - \mu|}{N}$. For a sample: $AD = \frac{\sum |X - \bar{X}|}{n}$, where the vertical bars denote absolute values.

What are the formulas for population variance and sample variance for ungrouped data?

Population variance: $\sigma^2 = \frac{\sum (X - \mu)^2}{N}$. Sample variance: $s^2 = \frac{\sum (X - \bar{X})^2}{n - 1}$.

How is the standard deviation defined mathematically, and what is its role among measures of dispersion?

Standard deviation ($\sigma$ for populations, $s$ for samples) is defined as the positive square root of the variance ($\sigma = \sqrt{\sigma^2}$, $s = \sqrt{s^2}$). It is the most widely used measure of dispersion.

What does the coefficient of variation measure, and what are its formulas?

The coefficient of variation ($V$) measures relative dispersion. For a population: $V = \frac{\sigma}{\mu}$. For a sample: $V = \frac{s}{\bar{X}}$. It is commonly multiplied by $100\%$ to be expressed as a percentage.

What is the relationship between the mean, median, and mode in symmetrical, positively skewed, and negatively skewed distributions?

In a symmetrical unimodal distribution: $\text{mean} = \text{median} = \text{mode}$. In a positively skewed distribution (longer right tail): $\text{mean} > \text{median} > \text{mode}$. In a negatively skewed distribution (longer left tail): $\text{mode} > \text{median} > \text{mean}$.

What is the formula for the Pearson Coefficient of Skewness ($S_k$)?

For a population: $S_k = \frac{3(\mu - \text{median})}{\sigma}$. For a sample: $S_k = \frac{3(\bar{X} - \text{median})}{s}$. For a symmetric distribution, $S_k = 0$.

How is skewness measured using the moments of a distribution?

Skewness is measured by dividing the third moment by the cube of the standard deviation: $S_k = \frac{\sum f(X - \mu)^3}{\sigma^3}$ for populations, or $S_k = \frac{\sum f(X - \bar{X})^3}{s^3}$ for samples.

What are the three curve classifications based on kurtosis shown below, and what do they describe?

Kurtosis refers to the peakness of a distribution curve: 1. Leptokurtic (a sharply peaked curve); 2. Mesokurtic (an intermediate, baseline curve); and 3. Platykurtic (a flat-topped curve relative to the mesokurtic curve).

What formula calculates kurtosis using the moments of a distribution?

Kurtosis is measured by dividing the fourth moment by the fourth power of the standard deviation: $\text{Kurtosis} = \frac{\sum f(X - \mu)^4}{\sigma^4}$ for populations, or $\text{Kurtosis} = \frac{\sum f(X - \bar{X})^4}{s^4}$ for samples.

What is the Sum Rule Principle in combinatorial counting?

If an event $E$ can occur in $m$ ways and a second event $F$ can occur in $n$ ways, and both events cannot occur at the same time, then event $E$ or $F$ can occur in $m + n$ ways. Extended to multiple mutually exclusive events, one of the events can occur in $n_1 + n_2 + n_3 + \dots$ ways.

What is the Product Rule Principle in combinatorial counting?

If an event $E$ can occur in $m$ ways and independent of this event, a second event $F$ can occur in $n$ ways, then combinations of $E$ and $F$ can occur in $m \cdot n$ ways. Extended to sequential events, all events can occur in the order indicated in $n_1 \cdot n_2 \cdot n_3 \dots$ ways.

How is the factorial function $n!$ defined, and what are the specific values of $1!$ and $0!$?

The factorial function $n!$ is the product of positive integers from $1$ to $n$ inclusive: $n! = n(n - 1)(n - 2) \dots 3 \cdot 2 \cdot 1$. By definition, $1! = 1$ and $0! = 1$.

What is a permutation, and what formula expresses the number of permutations of $n$ objects taken $r$ at a time?

A permutation is an arrangement of a set of objects in a specific, given order. The number of permutations of $n$ objects taken $r$ at a time is: $P(n, r) = {}^n P_r = \frac{n!}{(n - r)!}$.

What formula calculates the number of distinct permutations of a multiset containing repeated items?

For $n$ objects of which $n_1$ are alike, $n_2$ are alike, …, and $n_r$ are alike, the number of distinct permutations is: $P(n; n_1, n_2, \dots, n_r) = \frac{n!}{n_1! n_2! \dots n_r!}$.

What is the difference in calculation between ordered samples of size $r$ drawn with replacement versus without replacement?

With replacement (repetitions allowed), each draw offers $n$ choices, yielding $n^r$ ordered samples. Without replacement (no repetitions), the sample is an $r$-permutation yielding $P(n, r) = \frac{n!}{(n - r)!}$ ordered samples.

What is a combination, and what is its fundamental distinction from a permutation?

A combination of $n$ elements taken $r$ at a time ($C(n, r) = {}^n C_r = \binom{n}{r} = \frac{n!}{(n - r)! r!} = \frac{P(n, r)}{r!}$) is any selection of elements where order does not matter. The essential difference is that order counts in a permutation, but order does not count in a combination.