MAPC ASSIGNMENT 2023-2024
MPC-006 STATISTICS IN PSYCHOLOGY
SECTION – A
1. Describe the assumptions, advantages and disadvantages of non-parametric statistics.
A. Non-parametric statistics provide a valuable alternative to parametric methods, especially in
scenarios where the assumptions of parametric tests cannot be met or when dealing with data
of a nominal or ordinal nature.
Assumptions of Non-parametric Statistics
Non-parametric tests are particularly useful in the following situations:
1. Small Sample Size: One of the primary scenarios where non-parametric statistics come into
play is when dealing with a small sample size. Traditional parametric methods rely on
assumptions about the normality of the data distribution, which may not hold true for small
samples. In such cases, non-parametric tests provide a reliable alternative for statistical
analysis.
2. Doubtful Normality: In situations where the assumption of normality in the data
distribution is questionable, non-parametric tests offer a robust approach to analysis. This
is especially relevant when dealing with real-world data, which often deviates from the
idealized assumptions of parametric statistics.
3. Ordinal or Nominal Data: Non-parametric statistics are well-suited for data measured on
ordinal or nominal scales. These methods do not require the data to meet the interval or
ratio scale criteria, making them applicable to a wide range of research scenarios. Whether
the data can be expressed as ranks or categorized into distinct groups, non-parametric tests
offer an effective means of analysis.
4. Unknown Population Distribution: When the underlying distribution of the population is
unknown or not normal, non-parametric tests provide a versatile solution. Parametric
methods typically rely on assumptions about the population distribution, which may not
hold true in practice. Non-parametric tests, on the other hand, offer robustness in the face
of uncertainty regarding population characteristics.
Advantages of Non-parametric Statistics
Non-parametric methods offer several advantages over their parametric counterparts:
1. Suitability for Small Samples: Non-parametric tests are particularly useful when dealing
with small sample sizes. Traditional parametric methods may require larger sample sizes to
meet the assumptions of normality and homogeneity of variance. In contrast, non-
parametric tests are less sensitive to sample size, making them applicable even in situations
with limited data availability.
2. Fewer Assumptions: Non-parametric tests make fewer assumptions about the underlying
data distribution. This makes them more robust and versatile, especially when dealing with
, non-normal or non-parametric data. Researchers can use non-parametric methods with
greater confidence, knowing that they are not bound by stringent assumptions about the
data characteristics.
3. Relevance to Research Hypotheses: Non-parametric tests may be more appropriate for
certain research hypotheses. In cases where the assumptions of parametric tests are not
met or the data do not conform to traditional statistical assumptions, non-parametric
methods offer a more suitable alternative. Researchers can choose non-parametric tests
that align closely with the specific hypotheses under investigation, enhancing the relevance
and validity of their findings.
4. Treatment of Ranked Data: Non-parametric methods excel in handling data that are
inherently in ranks or categorical formats. Whether the data are expressed as ranks,
categories, or binary distinctions (e.g., good vs. bad), non-parametric tests offer an effective
means of analysis. This flexibility allows researchers to analyze a wide range of data types
without the need for complex transformations or assumptions about the underlying
distribution.
5. Ease of Learning and Application: Non-parametric tests are often easier to learn and apply
compared to parametric methods. The concepts underlying non-parametric statistics are
relatively straightforward, making them accessible to researchers with varying levels of
statistical expertise. Additionally, non-parametric tests typically involve fewer complex
assumptions and calculations, simplifying the analytical process for researchers.
Disadvantages of Non-parametric Statistics
Despite their advantages, non-parametric methods also have some limitations:
1. Wastefulness in Certain Scenarios: In situations where all the assumptions of a parametric
model are met and a parametric test could be used, employing non-parametric tests may be
considered wasteful. Non-parametric tests are generally less powerful than their parametric
counterparts, meaning they may require larger sample sizes to achieve comparable levels of
statistical significance. Using non-parametric tests unnecessarily in scenarios where
parametric methods are appropriate can result in inefficiencies in terms of statistical power
and sample size requirements.
2. Lower Power-efficiency: Non-parametric tests may have lower power-efficiency compared to
parametric tests when all conditions for parametric analysis are satisfied. The power-
efficiency of a statistical test refers to its ability to detect true effects or differences in the
data. In situations where parametric methods are applicable and assumptions are met,
parametric tests may offer greater power-efficiency compared to non-parametric tests. This
means that parametric tests may be able to detect smaller effect sizes or differences with a
smaller sample size compared to non-parametric tests.
3. Convenience Issues: Implementing non-parametric tests may be inconvenient due to the
scattered nature of required tables and resources. Unlike parametric tests, which often have
standardized procedures and readily available resources for calculation and interpretation,
non-parametric tests may require researchers to consult multiple sources and formats for the
necessary information. This can add complexity and time to the analytical process,
particularly for researchers who are less familiar with non-parametric methods.
, Non-parametric statistics offer valuable alternatives to parametric methods in situations where
assumptions cannot be met or when dealing with certain types of data. While they provide
flexibility and ease of application, researchers must carefully consider their suitability and
potential limitations in each analytical context.
2. Explain the concept of normal distribution. Explain divergence from normality.
A. The normal distribution is a cornerstone concept, portraying a symmetrical bell-shaped curve that
often characterizes various natural phenomena.
For example, when conducting a study on the heights of adult males, meticulously collecting data
from a representative sample. In a scenario where heights adhere to a normal distribution, the
majority of individuals cluster around the average height, yielding a bell-shaped curve that is
symmetric around the mean. This symmetrical distribution signifies that the mean, median, and
mode all converge at the center, illustrating a typical pattern observed in many real-world
phenomena.
Consider a hypothetical scenario where the heights of adult males in a population adhere to a
normal distribution with a mean of 175 cm and a standard deviation of 10 cm. This distribution
can be visually represented as follows:
Normal Distribution of Heights (FIGURE)
In this graphical representation, the x-axis represents height, while the y-axis indicates the
frequency or probability density. The curve illustrates a peak at the mean height of 175 cm,
gradually tapering off symmetrically on either side. This pattern underscores the typicality of a
normal distribution, where most individuals cluster around the mean height, with fewer outliers
at the extremes.
Divergence from Normality
MPC-006 STATISTICS IN PSYCHOLOGY
SECTION – A
1. Describe the assumptions, advantages and disadvantages of non-parametric statistics.
A. Non-parametric statistics provide a valuable alternative to parametric methods, especially in
scenarios where the assumptions of parametric tests cannot be met or when dealing with data
of a nominal or ordinal nature.
Assumptions of Non-parametric Statistics
Non-parametric tests are particularly useful in the following situations:
1. Small Sample Size: One of the primary scenarios where non-parametric statistics come into
play is when dealing with a small sample size. Traditional parametric methods rely on
assumptions about the normality of the data distribution, which may not hold true for small
samples. In such cases, non-parametric tests provide a reliable alternative for statistical
analysis.
2. Doubtful Normality: In situations where the assumption of normality in the data
distribution is questionable, non-parametric tests offer a robust approach to analysis. This
is especially relevant when dealing with real-world data, which often deviates from the
idealized assumptions of parametric statistics.
3. Ordinal or Nominal Data: Non-parametric statistics are well-suited for data measured on
ordinal or nominal scales. These methods do not require the data to meet the interval or
ratio scale criteria, making them applicable to a wide range of research scenarios. Whether
the data can be expressed as ranks or categorized into distinct groups, non-parametric tests
offer an effective means of analysis.
4. Unknown Population Distribution: When the underlying distribution of the population is
unknown or not normal, non-parametric tests provide a versatile solution. Parametric
methods typically rely on assumptions about the population distribution, which may not
hold true in practice. Non-parametric tests, on the other hand, offer robustness in the face
of uncertainty regarding population characteristics.
Advantages of Non-parametric Statistics
Non-parametric methods offer several advantages over their parametric counterparts:
1. Suitability for Small Samples: Non-parametric tests are particularly useful when dealing
with small sample sizes. Traditional parametric methods may require larger sample sizes to
meet the assumptions of normality and homogeneity of variance. In contrast, non-
parametric tests are less sensitive to sample size, making them applicable even in situations
with limited data availability.
2. Fewer Assumptions: Non-parametric tests make fewer assumptions about the underlying
data distribution. This makes them more robust and versatile, especially when dealing with
, non-normal or non-parametric data. Researchers can use non-parametric methods with
greater confidence, knowing that they are not bound by stringent assumptions about the
data characteristics.
3. Relevance to Research Hypotheses: Non-parametric tests may be more appropriate for
certain research hypotheses. In cases where the assumptions of parametric tests are not
met or the data do not conform to traditional statistical assumptions, non-parametric
methods offer a more suitable alternative. Researchers can choose non-parametric tests
that align closely with the specific hypotheses under investigation, enhancing the relevance
and validity of their findings.
4. Treatment of Ranked Data: Non-parametric methods excel in handling data that are
inherently in ranks or categorical formats. Whether the data are expressed as ranks,
categories, or binary distinctions (e.g., good vs. bad), non-parametric tests offer an effective
means of analysis. This flexibility allows researchers to analyze a wide range of data types
without the need for complex transformations or assumptions about the underlying
distribution.
5. Ease of Learning and Application: Non-parametric tests are often easier to learn and apply
compared to parametric methods. The concepts underlying non-parametric statistics are
relatively straightforward, making them accessible to researchers with varying levels of
statistical expertise. Additionally, non-parametric tests typically involve fewer complex
assumptions and calculations, simplifying the analytical process for researchers.
Disadvantages of Non-parametric Statistics
Despite their advantages, non-parametric methods also have some limitations:
1. Wastefulness in Certain Scenarios: In situations where all the assumptions of a parametric
model are met and a parametric test could be used, employing non-parametric tests may be
considered wasteful. Non-parametric tests are generally less powerful than their parametric
counterparts, meaning they may require larger sample sizes to achieve comparable levels of
statistical significance. Using non-parametric tests unnecessarily in scenarios where
parametric methods are appropriate can result in inefficiencies in terms of statistical power
and sample size requirements.
2. Lower Power-efficiency: Non-parametric tests may have lower power-efficiency compared to
parametric tests when all conditions for parametric analysis are satisfied. The power-
efficiency of a statistical test refers to its ability to detect true effects or differences in the
data. In situations where parametric methods are applicable and assumptions are met,
parametric tests may offer greater power-efficiency compared to non-parametric tests. This
means that parametric tests may be able to detect smaller effect sizes or differences with a
smaller sample size compared to non-parametric tests.
3. Convenience Issues: Implementing non-parametric tests may be inconvenient due to the
scattered nature of required tables and resources. Unlike parametric tests, which often have
standardized procedures and readily available resources for calculation and interpretation,
non-parametric tests may require researchers to consult multiple sources and formats for the
necessary information. This can add complexity and time to the analytical process,
particularly for researchers who are less familiar with non-parametric methods.
, Non-parametric statistics offer valuable alternatives to parametric methods in situations where
assumptions cannot be met or when dealing with certain types of data. While they provide
flexibility and ease of application, researchers must carefully consider their suitability and
potential limitations in each analytical context.
2. Explain the concept of normal distribution. Explain divergence from normality.
A. The normal distribution is a cornerstone concept, portraying a symmetrical bell-shaped curve that
often characterizes various natural phenomena.
For example, when conducting a study on the heights of adult males, meticulously collecting data
from a representative sample. In a scenario where heights adhere to a normal distribution, the
majority of individuals cluster around the average height, yielding a bell-shaped curve that is
symmetric around the mean. This symmetrical distribution signifies that the mean, median, and
mode all converge at the center, illustrating a typical pattern observed in many real-world
phenomena.
Consider a hypothetical scenario where the heights of adult males in a population adhere to a
normal distribution with a mean of 175 cm and a standard deviation of 10 cm. This distribution
can be visually represented as follows:
Normal Distribution of Heights (FIGURE)
In this graphical representation, the x-axis represents height, while the y-axis indicates the
frequency or probability density. The curve illustrates a peak at the mean height of 175 cm,
gradually tapering off symmetrically on either side. This pattern underscores the typicality of a
normal distribution, where most individuals cluster around the mean height, with fewer outliers
at the extremes.
Divergence from Normality
