The Buddha said that all beings are formless. The ultimate in data analysis can also be "no phase and no life".
Before talking about today's topic, I want to throw a conclusion:
"Doing data analysis, don't establish a mentality of grading yourself with the software you master, but you must use tools to avoid misleading career development!"
Why are we talking about the classification of data analysis tools today?
In fact, this involves a very common phenomenon in the workplace- tool contempt chain :
- Those who know python programming language, look down on those who use self-service BI tools for data analysis;
- Those who know how to use analysis tools like Tableau and FineBI, look down on those who know how to use data tools like SPSS and SQL;
- Those who can use SPSS and SQL, look down on those who use PPT for data reporting;
- Those who can use PPT for data reporting, look down on those who only use Excel for data statistics and sorting;
Inescapable chain of contempt
Not only in the data analysis industry, but also in industries such as programmers and product managers.
There are also many people and companies that follow the phenomenon of the contempt chain, and use this contempt chain to rank positions, or as a threshold standard for selection and recruitment.
And this contempt chain seems to live and die together with this post, coexist and coexist, and is inseparable:
For example, when I was new to data analysis a few years ago, I only learned to use Excel for data processing. All data cleaning, data processing and chart visualization work were done with Excel.
But seeing that many people on the Internet are learning Python programming, there was a saying "Python is the standard for data analysis" at that time, and I became a follower.
The boss who took me told me:
"On the Internet, it is purely fart to say that learning python can do data analysis . Python is indeed more professional than Excel for data analysis, but the real core is what data scenario and who is using it.
The data cleaning I do with Excel can definitely explode these tools, but I want to do some advanced visualization, and I will do it with python. Who stipulates that data analysts can only use one software?
After all, it is because of inferiority to the data analysis position . Data analysis has only appeared for a few years. Everyone thinks that using Excel to do data analysis is too low. It does not reflect our professionalism. If we do not learn a programming language, other Who in the department looks down on us? "
For my boss, I deeply agree that it is because of the misunderstanding of the data analysis job that has caused so many people to misunderstand data analysis tools
In fact, for so many years, excel, Tableau, FineBI, PowerBI, SPSS, Python... I have used almost every tool and software, and I don’t think which one is the most versatile
Tools have always been just tools. They are used to assist us in data analysis. The core key lies in people, that is, whether you really have data analysis thinking and ability
Like my former boss, using Excel is still awesome; if you don't have the ability, learning 10,000 languages is useless!
So many friends who are just getting started with data analysis bought a lot of Python, R language, VBA books and materials, and installed the environment of each language.
But in fact, the daily work of a data analyst is to use SQL access, Excel pivot charts, and BI analysis tools to drag and drop charts.
Therefore, knowing how to use tools is not ability. Understanding data thinking is the meaning of data analysts.
Tool is auxiliary
So back to our topic today, there are different data scenarios in data analysis, and a lot of tools will be used, so how to choose? Here are a few examples
1、excel
Don’t think that EXCEL only processes tables, you can use it as a database, you can also use it as an IDE, or even use it as a data visualization tool
It can create professional pivot tables and basic statistical charts, but due to the default settings of colors, lines and styles, it is difficult to create visual effects that look "tall"
Nevertheless, I still recommend you to use Excel
2, BI tool
The BI stars that have emerged in recent years, such as TB and qlk, have emphasized visualization, and have changed the traditional BI tools SAP BO and IBM's cognos (but in recent years they seem to be developing cloud BI)
I don’t talk about open source here. I haven’t seen BI that can be maturely applied. Mature BI tools such as FineBI (domestic) and Tableau (foreign) are highly recommended.
3、python
Friends who have studied Python data analysis know that among the visualization tools, there are many excellent tripartite libraries, such as matplotlib, seaborn, plotly, Boken, pyecharts, etc. These visualization libraries have their own characteristics and are also used in practical applications. Widely used by everyone
4 、 SPSS
SPSS is an introductory software for statistical analysis. If you want to get started quickly but don't want to learn programming, I recommend using SPSS
The focus of learning SPSS is not on the software itself, but on related statistical knowledge. This is also a foreshadowing of the previous suggestions, that is, you have to learn how to analyze "after inputting data, the software will show you the results"
I recommend a book: "The Implementation of SPSS/SAS EG for Data Processing Like a Tiger", written by Mr. Xu Xiaogang, very suitable for SPSS novices
What to say at the end
In fact, the topic discussed today is also one of the common phenomena in other industries. Since this kind of chaos has occurred, it shows that there must be reasons and logic for its occurrence. We don’t need to be too sensitive.
However, for our data analysts themselves, when you really do analysis, you will find that there are too many analysis tools on the market, and there are too many to master.
In fact, there is no need to dwell on this. Based on personal ability and the current data analysis environment, the applicable tools will naturally be selected. Finally, remember one sentence: Tools are used by people, and the focus is on people, not tools!