Video Lesson - Steps in data analysis

Loading...

Del curso dictado por Wesleyan University

Data Management and Visualization

673 calificaciones

Explorar los videos relacionados

En Coursera, encontrarás las mejores conferencias del mundo. Aquí te mostramos algunas de nuestras recomendaciones personalizadas para ti

Wesleyan University

Data Management and Visualization

673 calificaciones

Curso 1 de 5 en SpecializationData Analysis and Interpretation

Whether being used to customize advertising to millions of website visitors or streamline inventory ordering at a small restaurant, data is becoming more integral to success. Too often, we’re not sure how use data to find answers to the questions that will make us more successful in what we do. In this course, you will discover what data is and think about what questions you have that can be answered by the data – even if you’ve never thought about data before. Based on existing data, you will learn to develop a research question, describe the variables and their relationships, calculate basic statistics, and present your results clearly. By the end of the course, you will be able to use powerful data analysis tools – either SAS or Python – to manage and visualize your data, including how to deal with missing data, variable groups, and graphs. Throughout the course, you will share your progress with others to gain valuable feedback, while also learning how your peers use data to answer their own questions.

De la lección

Selecting a research question

We would like to welcome you to Wesleyan University's Data Analysis and Interpretation Specialization. In this session, we will discuss the basics of data analysis. Your task will be to select a data set that you would like to work with and to review available code books that help you develop your own research question. You will also set up a Tumblr blog that will allow you to reflect on these experiences, submit assignments and share your work with others throughout the course. First, you may want to check out the welcome video

Welcome Video5:38

Video Lesson - Steps in data analysis7:20

Conoce a los instructores

Lisa Dierker
Professor
Psychology

[MUSIC]

Statistics plays a significant role across the physical and social sciences.

And it's arguably the most salient point of

intersection between diverse disciplines.

>> Statistics is the common language of science.

Scientists use statistics to convert data into useful information.

This is exactly what you will do in this course.

Through your own original research,

you will convert existing data into useful information.

You use data to tell a story that's of interest to you, and of value to others.

Statistics is there for a process, where we are collecting data,

summarizing data, and interpreting data.

The process of statistics starts when we identify what group we wanna study or

learn something about.

We call this group the population.

The word population, and

throughout its entire course, is not just used to refer to people.

It is used in a more broad statistical sense.

Where population refers to an entire group you want to focus upon.

This can be an entire group of people or animals or

insects, or inanimate objects like apartment buildings or craters on Mars.

[MUSIC]

>> For example, we might be interested in the opinions of the population of US

adults about the death penalty.

[MUSIC]

How the population of mice react to a certain chemical.

[MUSIC]

The average price of the population of all one bedroom apartments in a certain city.

Population, then, is the entire group that's the target of our interest.

In most cases, the population is so large, that as much as we want to,

there's absolutely no way that we can study all of it.

A more practical approach would to be to examine and

collect data only from a sub-group of the population, which we call a sample.

We call this first step which involves choosing a sample and

collecting data from it, producing data.

Since for practical reasons we need to compromise and examine only a subgroup of

the population rather than the whole population, we should make an effort to

choose a sample in such a way that it will represent the population as well.

[MUSIC]

James Thompson has been studying pollination success of an obscure flower

that blooms at high altitudes, the glaciered lily.

We're looking at a high altitude species.

And as people interested in global change and

climates have realized this is where we might first see

things like species doing poorly because the situations have changed,

and in fact there's a lot of focus on high elevation research in

the general context of climate change.

>> There are millions of these flowers across the Rockys.

And Professor Thompson can not study them all, he has to choose a sample.

>> If I'm to say anything accurate those data have to really reflect

what's happening out here.

So for instance when I look at a sample of flowers I have to be thinking

all the time about whether I'm selecting a set of flowers that's

adequately representative of the whole of what I wanna talk about.

[MUSIC]

>> This is true whether we're studying flowers, craters on Mars, or

the opinions of US adults regarding the death penalty.

Our sample would not represent US adults, if we only asked Republicans or

only asked Democrats.

Such a sample would not represent the population.

Data sets can be very different depending on what is being studied.

This data can take the form of answers to survey questions,

tables of numbers such as crater details,

or in the case of glacier lilies, observations collected over many years.

Essentially, I've asked a very simple question.

What causes a flower to be a success?

Does it get pollinated?

Does it set a fruit?

Does it make seeds?

How many seeds does it make?

>> To make sense of this data, it needs to be summarized in a meaningful way.

This is called exploratory data analysis.

Exploratory data analysis often reveals new ways to think about the data.

>> As frequently happens in science,

the more carefully I looked, the more things I saw to interest me.

>> Exploratory data analysis helps scientists refine their questions.

And sometimes even reveal entirely new questions.

>> If climates are changing, relationships between plants and

pollinators and other mutually beneficial relationships, those may be disrupted.

>> Scientists studying climate change have often wondered what effect

a warming climate will have on the relationship between plants and animals.

Is it possible that small changes in climate could

have a big impact on these relationships?

Exploratory data analysis suggests that that is a question Professor Thompson

might be able to answer using 30 years of data.

This leads up to the final step, inference.

What can we infer about the population as a whole from the data in our sample?

Remember, after exploratory data analysis,

we're able to ask specific questions of our data.

Inference is where we come full circle with the hope of revealing new knowledge

about the population.

So what can Professor Thompson infer about Glacier Lilies?

What his data revealed is that glacier lilies and

the bees that pollinate them are becoming separated in time.

As the climate warms, lilies bloom sooner before the bees arrive.

>> My paper on glacier lilies, as far as I can tell,

is the first demonstration of it or the first even plausible demonstration of it.

It's not an easy thing to show.

It's an easy thing to say, hey, this could happen.

My long term data sets have allowed me to do is to say, yeah, and

it looks like it has happened.

>> James Thompson was able to explore his data to show how

climate change causes plants and animals to become disconnected in time.

You too will be looking at large data sets and

asking new questions of interest to you.

You won't be creating new data, but you will create new knowledge through

exploratory data analysis and inferential data analysis.

Statistics education is most often conducted

within a discipline specific context or as generic mathematical training.

Our goal is instead to create meaningful dialogue across disciplines.

Ultimately, this experience will help you on your way to engaging in

interdisciplinary scholarship at the highest levels.

Explora nuestro catálogo

Inscríbete de manera gratuita y obtén recomendaciones personalizadas, actualizaciones y ofertas.

Coursera brinda acceso universal a la mejor educación del mundo, al asociarse con las mejores universidades y organizaciones, para ofrecer cursos en línea.

© 2019 Coursera Inc. Todos los derechos reservados.

Descargar en la App Store

Consíguelo en Google Play