What is the difference between Descriptive Statistics and Inferential Statistics?

Summary

This video from 'Psy Versus Psy' demystifies the role of statistics in psychology, focusing on the distinction between descriptive and inferential statistics. Descriptive statistics, such as mean, median, mode, and standard deviation, summarize and describe data, whereas inferential statistics use probability to make inferences about populations. The channel uses an engaging example of a study on natural versus artificial environments to illustrate how these statistical methods help interpret data and answer research questions.

Takeaways

• 📊 Descriptive Statistics: These are used to summarize and describe the data collected from a sample, including measures like averages, medians, modes, variance, standard deviation, minimum, and maximum.
• 🔍 Inferential Statistics: This type of statistics uses descriptive statistics and probability theory to make inferences about a population from sample data, determining if observed differences are significant or due to random chance.
• 🌿 Example Study: The script discusses a hypothetical study comparing preferences and stress levels in natural versus artificial environments using measures like usage and salivary cortisol levels.
• 🧪 Salivary Cortisol: The study example mentions using saliva samples to measure cortisol, a stress hormone, to assess stress levels in different environments.
• 📝 Data Analysis: The script explains the process of analyzing data, starting with descriptive statistics to understand the sample before moving to inferential statistics to make conclusions about the population.
• 📉 Statistical Tests: It mentions various inferential statistical tests such as t-tests, ANOVAs, chi-squared tests, correlations, and regressions that are used to analyze differences and relationships in data.
• 🎯 Key to Inferential Statistics: The presence of measures of probability like p-values or Bayes factors indicates inferential statistics, which go beyond describing the data to make inferences about the population.
• 🔑 Making a Determination: To identify if a statistic is descriptive or inferential, one should look for measures of probability and whether the statistic is making comparisons or drawing conclusions about the population.
• 🤔 Understanding Statistics: The script aims to demystify statistics, emphasizing their importance in psychological research and the scientific process.
• 🔑 Importance of Both Types: Both descriptive and inferential statistics are crucial for understanding and interpreting data in psychology studies.
• 👍 Encouragement: The video script encourages viewers to engage with the content by liking and subscribing for more psychology-related videos.

Q & A

What is the main topic of the video script?

The main topic of the video script is the difference between descriptive statistics and inferential statistics, with an example related to the preference for natural versus artificial environments and their impact on stress levels.

Why are statistics important in psychology research?

Statistics are important in psychology research because they are a crucial part of the scientific method, helping to analyze and interpret data, and to make inferences about populations based on sample data.

What is an example of a natural environment used in the script?

An example of a natural environment used in the script is an outdoor area with tables and chairs surrounded by plants and vegetation where people can eat their lunch.

What psychological literature evidence is mentioned in the script regarding natural environments?

The script mentions evidence from psychological literature that suggests people have a preference for natural environments over artificial ones and that exposure to natural environments can reduce stress.

What is the role of descriptive statistics in the given research example?

Descriptive statistics in the research example are used to summarize and describe the data collected, such as the average number of people visiting each location per day, the median, mode, and standard deviations of visitor numbers.

What is the purpose of inferential statistics in the context of the research?

The purpose of inferential statistics in the context of the research is to determine whether the observed differences between the natural and artificial environments are statistically significant and not just due to random chance.

What is an example of an inferential statistical test mentioned in the script?

An example of an inferential statistical test mentioned in the script is the t-test, which can be used to compare the usage of the two sites over the week.

How does the script differentiate between descriptive and inferential statistics?

The script differentiates between descriptive and inferential statistics by stating that descriptive statistics describe the sample data, while inferential statistics use probability measures and reasoning to make inferences about the population.

What is the significance of cortisol levels in the research example?

Cortisol levels are significant in the research example because they are a measure of stress hormones, and any differences in cortisol levels between the natural and artificial environments can indicate differences in stress levels.

What are some key indicators that a statistic is inferential according to the script?

According to the script, key indicators that a statistic is inferential include the presence of measures of probability like p-values or Bayes factors, and whether the statistic is making comparisons or drawing conclusions about populations beyond the sample data.

How does the script suggest enhancing the understanding of statistics?

The script suggests enhancing the understanding of statistics by demystifying some of the concepts that may initially seem intimidating and by providing clear examples and explanations of how these statistics are used in research.

Outlines

00:00

📊 Descriptive vs. Inferential Statistics

This paragraph introduces the topic of the video, which is the difference between descriptive and inferential statistics. The speaker uses the context of a psychology channel to explain why statistics are essential in scientific research, even though they might seem intimidating at first. The example of a study comparing the usage and stress levels of people in natural versus artificial environments is given to illustrate how these statistical concepts apply in real-world research scenarios. Descriptive statistics, such as means, medians, modes, and standard deviations, are explained as measures that describe the data collected from the sample.

05:00

🔍 Understanding Inferential Statistics

The second paragraph delves deeper into inferential statistics, explaining how they are used to go beyond the data at hand to make comparisons and draw conclusions about populations. The speaker discusses the importance of measures of probability, such as p-values or Bayes factors, as indicators of inferential statistics. The paragraph also highlights the distinction between simply describing data (descriptive statistics) and making inferences about the population (inferential statistics). The video aims to demystify statistics and encourage viewers to appreciate their role in psychology and scientific research.

Keywords

💡Descriptive Statistics

Descriptive statistics are numerical measures that summarize and describe the features of a set of data. In the video, they are used to calculate averages, medians, modes, and standard deviations to describe the data collected from the two different environments. They help in understanding the central tendency and variability within the sample data, which is essential for presenting the findings of the research.

💡Inferential Statistics

Inferential statistics are used to make inferences about a population based on sample data. The video explains that these statistics allow researchers to determine if the observed differences between groups are statistically significant or due to random chance. Examples of inferential statistics mentioned include t-tests, ANOVAs, chi-squared tests, correlations, and regressions.

💡Psychology

Psychology is the scientific study of the human mind and its functions, particularly those affecting behavior. The video is from a channel dedicated to psychology, which discusses the application of statistics in psychological research. The script uses a psychological study on preferences for natural versus artificial environments to illustrate the use of statistics.

💡Natural Environments

Natural environments refer to settings that are characterized by the presence of plants, vegetation, and other natural elements. The video script mentions that there is psychological literature suggesting people prefer natural environments, which can also reduce stress levels, as part of the research context.

💡Artificial Environments

Artificial environments are man-made settings, often characterized by buildings and concrete without much natural vegetation. In the video, these are contrasted with natural environments to explore preferences and stress levels, using them as a control in the hypothetical study.

💡Salivary Cortisol

Salivary cortisol is a measure of the stress hormone cortisol found in saliva. The script describes collecting saliva samples to measure cortisol levels as a way to assess stress in the study comparing natural and artificial environments.

💡ELISA Assay

ELISA (Enzyme-Linked Immunosorbent Assay) is a biochemical test used to detect and quantify substances such as proteins or hormones. In the video, an ELISA assay is mentioned as the method for measuring salivary cortisol levels in the research study.

💡Stress

Stress is a psychological and physiological response to demanding or threatening situations. The video discusses how exposure to natural environments can reduce stress levels, and cortisol is used as a biomarker to measure stress in the study.

💡T-test

A t-test is a statistical method used to determine if there is a significant difference between the means of two groups. The video script mentions using a t-test to compare the usage of the natural and artificial areas to see if there is a preference based on the number of visitors.

💡Correlation

Correlation is a statistical term that refers to a measure that expresses the extent to which two variables are linearly related. The script suggests finding a correlation between cortisol levels and the time spent in the natural area, indicating that longer stays might be associated with lower stress.

💡P-values

P-values are used in inferential statistics to determine the probability that the observed results occurred by chance. The video explains that the presence of p-values is an indicator that the statistics being discussed are inferential, helping to assess the significance of the findings.

💡Bayes Factors

Bayes factors are a measure used in Bayesian statistics to compare the probability of two competing hypotheses. The video briefly mentions Bayes factors as another type of measure of probability used in inferential statistics to assess the strength of evidence for a hypothesis.

Highlights

Statistics are an integral part of scientific research in psychology.

Descriptive statistics summarize and describe the data from a sample.

Inferential statistics make inferences about populations based on sample data.

Means, medians, modes, and standard deviations are examples of descriptive statistics.

Inferential statistics use probability to determine if observed differences are significant.

The video uses an example of studying natural versus artificial environments to illustrate the concepts.

Salivary cortisol levels are measured to assess stress differences between environments.

Descriptive statistics describe the data for each area, such as average visitor numbers.

Inferential statistics determine if the differences between natural and artificial areas are statistically significant.

T-tests, ANOVAs, chi-squared tests, and regressions are examples of inferential statistics.

Probability measures like p-values indicate inferential statistics.

Descriptive statistics simply describe the data, while inferential statistics make broader comparisons or conclusions.

Both descriptive and inferential statistics are crucial for understanding and interpreting research data.

The video aims to demystify the intimidating aspects of statistics for psychology students and professionals.

The channel 'Psy vs Psy' provides content on various psychological topics, including statistics.

The video explains the practical application of statistics in a research context, using natural and artificial environments as a case study.

The importance of distinguishing between descriptive and inferential statistics for accurate data analysis is emphasized.

Transcripts

what's the difference between

descriptive statistics and inferential

statistics stay tuned

[Music]

welcome to psy versus psy the channel

for all things psychology so if that's

what melts your butter stick around you

might enjoy a few of our other videos

now some of you might be wondering why a

psychology channel would be talking

about statistics after all nobody needs

the creepy specter of math hanging

around to spoil their knock it off

spoil their psychology fun but the

reality is that statistics are an

important part of science and so on

occasion we try to help demystify some

of the arcane magic that can be

intimidating at first glance today we're

talking about the difference between

descriptive statistics and inferential

statistics now whether you're doing

research on your own or just consuming

research so you can use evidence-based

practices to improve what you do knowing

how these statistics are used is

important now we could talk about this

in the abstract sense but i always like

to use an example experiment as sort of

a starting point

at my work there's a few little outside

areas that have some tables and chairs

so people can eat their lunch when the

weather's nice i find some of these

areas especially appealing because

they've got a lot of plants and

vegetation around

that's much better to me than just

buildings and concrete

now there's some evidence in the

psychological literature that people

have preferences for more natural

environments than artificial

environments and that exposure to

natural environments can even reduce

stress

so if i wanted to do a study i could

explore the differences between these

two spaces

i could look at preferential usage

measuring how many people use each space

on average per day over the course of a

week i could also ask people in each

space to give me a saliva sample

that's not weird right because it's for

science if i collect saliva samples i

can do an elisa assay to measure

salivary cortisol now cortisol is a

stress hormone so any differences in

cortisol levels between the two

locations natural and artificial are

likely to be differences in stress

levels okay so let's say i collect this

data and now i have to analyze it the

first thing i might do is look at some

averages how many people visited each

location per day on average across the

week

or i could look at the median or maybe

the mode

maybe i want to account for how variable

the number of visitors were across days

so i calculate the standard deviations

now we have videos with more detail on

how to calculate those things and what

they are which i'll link to in the

description but of all of these

statistics do the same thing they

describe the data collected for each of

the two areas and so we call these

descriptive statistics now that makes

sense

means medians modes

variance standard deviation minimum

maximum all of those things would be

considered descriptive statistics they

just describe the sample that we

collected

now those numbers are useful but the

heart of my research question is whether

i can be confident that the

numbers that i got for the natural area

and the artificial area are different

from each other or not

after all if i flip a coin ten times i

might expect to get exactly five heads

and exactly five tails but if i got six

heads and four tails that doesn't mean

the coin is biased

random chance could cause some

variability in the measurement

so in order to tell whether the

differences between the two conditions

are reliably different from each other

or if it's just due to random chance i

need to do some kind of statistical test

statistical tests are generally what we

call

inferential statistics because we're

using the descriptive statistics and

laws of probability to infer or deduce

based upon evidence and reasoning

whether this difference is significant

and we should pay attention to it or not

so if i do a t-test to compare the usage

of the two sites over the week i might

find that people preferred to use the

more natural area

or maybe i find that the cortisol levels

of people in the natural area are

inversely correlated with the amount of

time spent there that is the longer

they'd stayed in that area the less

stressed they were

so things like t-tests anovas

chi-squared tests correlations

regressions these are all examples of

inferential statistics now when i'm

trying to figure out if a statistic is

descriptive or inferential there are a

couple of things that i can look at to

help make that determination

one key thing to look for is whether

there are measures of probability like

p-values or bayes factors if you've

heard of those or some other estimate of

confidence in the inference about the

differences between the groups and that

would be a clue that i'm looking at

inferential statistics if there's some

sort of measure of probability then

you're probably looking at inferential

statistics the second thing is whether

this statistic is simply describing the

data at hand or is it going beyond the

data a little bit to make comparisons or

draw conclusions about the populations

from which these two different

conditions are drawn

if it's making comparisons it's likely

to be an inferential statistic

you're likely to run across both

descriptive and inferential statistics

in any psychology study and both are

important for helping us understand and

interpret our data i hope this video has

made statistics just a little bit less

scary

if you found this video helpful hit the

like button subscribe to get more videos

on all things psychology and until next

time

keep thinking

[Music]

i don't think we're quite ready for you

yet

[Music]