当前位置：网站首页>Full nanny tutorial of Market Research Competition (experience sharing)

Full nanny tutorial of Market Research Competition (experience sharing)

2022-06-25 12:21:00 【Halosec_ Wei】

2017 In, I participated in the Zhengda Cup College Students' Market Research and analysis competition , Got the second grade in the country （ It took 1200 Oceania's national tournament , Love dearly ）, Recognized by the Ministry of education 、 Can guarantee the research competition class to add points almost 5 Points can be filled A Class competition , See that there is no more complete online competition experience , I think of the pit I stepped on , Today, I would like to share with you the experience of all nannies in the city survey competition .

Directory as follows , Total words 2W word , About reading 2 In hours or so , You can take notes , among Individual competition and questionnaire data analysis tutorial pure dry goods , After reading it, you will get a provincial award or simply , Typing is not easy to , Everyone thinks it's useful after reading it. Remember to praise Sanlian , Be deeply grateful .

Catalog

1、 Introduction to the market research competition
2、 The time planning process of the city survey competition
3、 Personal knowledge contest （ Online examination ） strategy （480 Page question bank sharing ）
4、 Team practice competition
4.1 How to form a team , Division of labor ？
4.2 Determine the topic and design the questionnaire ？（28 Video sharing of the defense of the first prize of the national competition ）
4.3 Distribute and collect questionnaires
4.4 Questionnaire data processing and analysis ( dried food )
4.4.1 Questionnaire data analysis process
step1： User profile analysis ： The sample is representative
step2： Reliability and validity test ： The survey tool is stable
step3： Descriptive statistics ： Sample characteristic behavior analysis
step4： Inferential statistics ： Research core

4.4.2 Questionnaire data analysis 4 Big frame
Affect the research framework of relationship
Policy status research framework
Adjust the / Intermediary research framework
Research framework of experimental differences

4.5 Write reports and PPT
4.6 Competition defense skills
4.7 Share a questionnaire data analysis artifact

1、 Introduction to the market research competition

National undergraduate market survey and analysis contest （ And the cross-strait college students' market survey and analysis competition in mainland China ） By the Chinese Society for business statistics 、 Sponsored by the Teaching Steering Committee of statistics major in Colleges and universities of the Ministry of education .

Competition form ： The competition is divided into two categories: undergraduate group and junior group , There are two competition links: knowledge competition and practice competition .

Knowledge competition is an individual competition , Take online examination , It mainly assesses students' understanding of basic theories and basic knowledge 、 Mastery of skills .
Practice competition （ Each team consists of 3-5 It's made up of two players ）, The qualified players of the knowledge competition form a team to participate in the practice competition , The practice competition contains written reports 、 The on-site defense consists of two parts , It mainly inspects students' ability to combine theory with practice 、 The ability to solve practical problems .

Competition influence ： The national competition was founded in 2010 year . The cross-strait race begins with 2012 year , It was officially listed as the educational exchange program of the Ministry of education to Taiwan . This calendar is 7 Months , Pass the school competition 、 Provincial competition 、 The national competition and other multi-layer on-site display defense plus comprehensive score of report review , In addition, students come from different majors and other factors , Be able to cultivate students' teamwork 、 Live show 、 Language expression , And the comprehensive application ability of various information technologies , Multiple perspectives 、 Improve students' comprehensive quality at multiple levels .

Participants ： Full time junior college 、 Undergraduate 、 Master's degree students can apply , There is no limit to majors .

Competition organization ： Market Research and Analysis Competition Organizing Committee , It has a secretariat 、 The review team and the sub division organizing committee . The Secretariat is responsible for the organization and implementation of professional competitions . The judges of the competition are college teachers 、 It is composed of professionals from government statistics departments, professional survey companies and senior personnel from market research departments of large enterprises . There are several sub competition areas in the competition , Colleges and universities that have applied for approval to set up division areas , Set up sub division group meetings , Be responsible for the organization of the undergraduate group events in the competition area （ There are no divisional competitions in Higher Vocational and technical college groups , Each school selects excellent teams to directly participate in the national finals ）.

Registration method ： Take the school as a unit , The relevant departments of each school shall organize the registration of the school .

Official website ：http://www.http://china-cssc.org

2、 The time planning process of the city survey competition

The first level ： Personal knowledge contest , Form of personal online test , Every year, 11 Mid to 12 month Mid , Each candidate will take one or two examinations according to their own conditions . PS： This thief pit , At that time, I almost did not , It's a lot of money to find the answer .

The second level ： School competition , The form of group practice competition ,3 By the end of the month . The players who pass the personal online test form a team by themselves , The participating institutions will organize the selection within the school , Recommend excellent teams to participate in provincial trials .

PS： At that time, I was cramming , The analysis only started one week before the official competition , Fortunately, the teammates want to win the prize and guarantee the research , Still serious （qiangbi） It's done .

The third level ： Provincial competition , 4 By the end of the month , Organized by the sub division organizing committee , Select the winning team to participate in the national finals according to the unified competition rules and scoring standards formulated by the competition organizing committee .

PS： At that time, our judges looked at the novelty and PPT、 Report format , Not professional at all

The fourth level ： The national finals （ And the mainland area trials of the cross strait university students' market survey and analysis competition ）,5 At the end of the month , The competition selected 6 An undergraduate team ,1 A specialist team represents the teams from the mainland , Participate in the finals of the cross-strait college students' market survey and analysis competition .

PS： National competition fee 1200 Dayang school doesn't give reimbursement ...

The fifth level ： The final of the university students' market survey and analysis competition across the Taiwan Straits , On 8 In the last ten days of the month, it will be held in turn in the mainland and Taiwan .

3、 Personal knowledge contest （ Online examination ） strategy

The result of the individual knowledge competition passed 60 A pass , Before the school competition 30% The contestants passed the computer test , That is to say, did you get 60 It's possible to pass the computer test , Be sure to get ahead 30%, So the computer test must be fully prepared , We must pay attention to , But you don't have to worry , The answer Lord listed the title document in the following text , Remember to save , When answering questions, you can search directly , Otherwise, you can answer the question of doubting life , At that time 60 More than a dollar to buy the members of various answer websites .

Special notes

After signing up for payment （30 Yuan ）, You can take a mock exam online , But there is only one chance ;
The topic type is single choice , multi-select , Judgment questions , Total score 100 branch ;
The examinee starts timing after entering the test page , At the end of the answer , Candidates can hand in their own papers , The examination paper will be handed in automatically as soon as the examination time arrives . After the examinee hand in the paper , Show scores directly ;
Each person can take the formal exam twice , First pass , No need to take the second exam ; Students who fail the first exam , You can take the exam a second time , The final score is the higher of the two examinations , Many people can work together to do questions . Pay attention to is ： Everyone has two chances to answer the questions ;
After the contestants enter the online examination system , By user name （ My name ）、 cell-phone number （ Be sure to be consistent with the one submitted during online registration ） And mobile phone verification code （ Send SMS to my mobile phone ）, You can successfully log in ;
Please take the test during the whole test , Use the same browser to answer questions , Do not close or switch pages during answer , If the examinee does not use the same browser to answer questions or switch other pages during the examination , It will cause the loss of test data . The resulting losses , The test organizer will not assume any responsibility ;
Knowledge contest （ Personal online test ） The contestants who have passed the examination may voluntarily apply for the certificate of Professional skills in national market research and analysis （CRA） certificate （ Need to hand in 120 ocean , But having a certificate is better than not having one ）

The outline of knowledge contest

There is no designated reference book in the knowledge contest , The purpose is to test the basic statistical knowledge and market research knowledge of the contestants . The test is divided into two categories: Undergraduate test questions and higher vocational test questions , Each type of test includes two parts: statistical knowledge and market research knowledge .

The specific contents are as follows ：

Scope of undergraduate statistical test ：

The scope of statistics test in higher vocational colleges ：

Scope of market research knowledge （ Do not distinguish between undergraduate and junior college ）

research 、 Summary of statistical knowledge exercises

When reviewing statistics and market research, I sorted out some question banks , Full version 428 page , There are too many to list , Students in need can take it by themselves ：

research 、 Summary of statistical knowledge exercises （ Online examination question bank + answer ）

4、 Team practice competition

4.1 How to form a team , Division of labor ？

After the personal knowledge competition , Don't rush to choose a topic , We need to form a team , The general practice team can be composed of 3~5 Composed of , The general roles of the division of labor are ：

captain ： Strong sense of responsibility , Good at communication , Responsible for overall planning 、 Follow up the game ,PPT Strong production ability ;

Roadshow contestants ： It's better to be a girl , Good temperament is required , The mouth is like a flower ;

Data analysis players ： Data processing , analysis 、 modeling , Strong data analysis ability is required ;

Chinese modeling player ： Descriptive statistics 、 Strong clerical skills , Be able to put simple words into professional and complex ,

Meitu XiuXiu contestant ： Students majoring in design , Responsible for cover making , Video clip , Graphic design .

Division of labor for field investigation of the whole staff , Revise the full text and conclusion together , Suggestions, etc .

The captain is the key , Is the captain responsible for ensuring that your game can be completed completely , At that time, we just chose a senior student as our team leader , But he has been very busy , Almost forgot , I temporarily acted as the captain , But when you get into shape , The senior student immediately played his role .

The roadshow contestant decides the score of the on-site defense , Just one minute on stage , Ten years' work under the stage , Face project must be done well , The judges are also human , People will have aesthetics , So a good-looking sister roadshow can add a lot of points , Of course, the most important thing is the novelty of the content and title of the article .

In fact, the programming modeling players mainly analyze the questionnaire data , This is mainly analytical thinking , Later, I will explain the tutorial of questionnaire data analysis , In fact, Xiaobai can also learn , Recommendation tool ：

SPSSPRO： Free online data analysis platform

The model tutorial will teach you how to operate , Text version and video version are available for you to choose from , Practice from the case 、 Model theory, and so on , Xiaobai from 0 You can also understand data analysis ！

Covering all professional statistical algorithm models , Continuous updating , So that your various algorithm requirements can be met one-stop
One key generation analysis and interpretation , Automatic annotation of key terms , Can read analysis report without statistical knowledge

4.2 Determine the topic

Don't rush to choose a topic ！ Don't rush to choose a topic ！ Don't rush to choose a topic ！

Important things are to be repeated for 3 times .

More important than topic selection , Is to see how others won the prize first , Here, the respondent has prepared a collection of previous excellent works of Zhengda cup , And the live defense video of the final .

I collected it 《 Zhengda Cup National College Students' Market Research and analysis competition - Excellent works 》 Video collection . If necessary, you can follow wechat official account SPSSPRO Get a reply 【 City Survey contest 】 obtain .

First of all, read and study the research reports of national awards in recent years . Don't look at works below the national level , It's easy to be biased , Do more harm than good . These national award-winning works are the upper limit of the city survey competition , It has great reference value . Topics are divided into enterprise proposition and independent proposition ：

Enterprise proposition

After the school game , No race saved , The relevant organizing committee will directly select and promote the national competition , What we pay attention to is that the research method is more scientific , The conclusion has more commercial value . Relative to autonomous proposition , Both risks and competition are fierce , Although we can go directly to the National Games step by step , But there are fewer places , It's easy to brush , Unless you are confident in your own and your team's research expertise .

PS： Received the latest news , The enterprise proposition has no advantages this year , Just like autonomous proposition School games Provincial competition National Games .

Autonomous proposition

The key to success is novelty , A proposition that requires more practical significance 、 The easier it is to win a prize for a novel proposition , It is recommended for novices to attend for the first time . Generally, if we are not very professional research er It is better to choose independent proposition , You can get at least a provincial award if you make a decision . When determining the topic , Try to avoid the research topics written by predecessors , Find hot topics or new topics , For example, the metauniverse is very hot recently , Then we can write about this , Then we should ensure that our research results have commercial or academic value , If the research results have no practical significance , In the end, it's hard to get high awards , Try to match the theme of the main melody .

How to design a questionnaire ？

Confirm the selected topic , Next is the design topic . First of all, we must deeply realize the importance of the questionnaire , Questionnaires play a central role in data collection and analysis , The main factors affecting data quality . Designing a reasonable questionnaire needs to look at problems from multiple perspectives , The questionnaire is designed according to the research ideas and contents . The purpose and filling requirements of the questionnaire are stated at the beginning of the questionnaire , Problem design is the basic information first , The second is the investigation of the cognitive situation related to my own research direction , After that, the attitude and view of the research direction events are investigated , Thank you again at the end . After completing the questionnaire, send it to the group for discussion , Carry out pre investigation after the first edition is obtained through further modification , Process the data collected from the pre survey 、 analysis 、 explain , Further modification of the inappropriateness , Formal investigation can be carried out after modification and review . Before you start working on the questionnaire , We need to figure out some more core issues first .

Who are the target respondents ？

We need to have some understanding of the target group , Understand the customer group portrait of the research group , To ask the right questions .

What is the purpose of the study ？

List the objectives and needs of the study , For example, I would like to know whether there are differences in the evaluation of certain things by men or women in the research group ？ So let's list two questions ： The gender of the research group and the evaluation of the event , By analyzing these two problems , To verify our research objectives and assumptions , The specific problem design idea can be ： Purpose → hypothesis → Variable → problem Proceed from the fundamental purpose , We can derive some specific assumptions . If conditions permit , You can interview the target group first , Help us expand our thinking and form assumptions . Each specific hypothesis reveals which variables we should measure , Each variable may need to be measured with multiple questions .“ Purpose → hypothesis → Variable → problem ” This structure tree helps us to think about what problems should be raised according to the research objectives . You can also collect the questions the team wants to know by brainstorming with team members （ There is no need to write a standard questionnaire ）. But here's the thing , Every problem needs a reason to exist , That is, its contribution to the research objectives . Try not to leave too many questions in the end .30 A questionnaire with three questions will probably cost users 20 Minutes to fill in , If exceeded 20 minute , The response rate of the questionnaire and the quality of the answers will decline .

What should be paid attention to in the questionnaire design is ：

Be sure to consider the language ;
The questionnaire was designed with a five level scale ;
The setting of options should be neutral , There can be no hint , Do not guide selection ;
The design of the problem should be targeted , The wording of the question must be clear and unambiguous , Incorrect wording may lead to wrong conclusions and inappropriate analytical conclusions

4.3 Distribute and collect questionnaires

The questionnaire website is highly recommended here , Because the questionnaire network can directly import the questionnaires online after collecting them spss Analysis platform spsspro, It's completely free , And there are many free questionnaire design templates , One click to quickly collect questionnaires .

Free templates

【 Free templates 】 It is a quick design tool prepared by the questionnaire network for users . You can find the entry at the top of the project management page , Enter the special area .

There are many ways to find templates in the template zone ：

► Keyword search

Enter the project keywords you want to find , You can get template recommendations with keywords , It is a relatively fast operation .

► Special section

in addition , The questionnaire network aims at the project scenarios commonly used by users , We have prepared the corresponding special topic sections , Put all the templates related to a specific scene together , It is more convenient for users with this requirement to browse .

such as , During the epidemic , The questionnaire network is specially set up 【 Special topics on epidemic prevention and control 】, Provide help for the epidemic prevention and control work of all walks of life .

► Tag search

There is also a search method for template libraries , It is to search according to different types of tags . at present , The questionnaire network supports “ Project type ”、“ scene ” as well as “ Industry sector ” Search by category , There are a large number of templates to choose from under each tag .

When you find your favorite template , You just need to open , And click... In the upper right corner of the template “ Reference the template ”, It will appear on your project page . You just need to edit and publish on the project page .

Question bank

When you find the template , But you may need to modify or add some topics ; Or, , You haven't found a particularly suitable template , I hope I can do some topic design by myself and I have no way to start , Questionnaire network “ Question bank ” The function will be able to show your skills .

You can find it on the left side of the project edit page ：

Click on the question bank , You can see the classification of different topic types , Meet different problem creation needs ：

for instance , If you want to write a questionnaire about employee satisfaction , You can choose “ human resources ” classification , Then you can see the related topics recommended by the question bank , Follow the instructions , You can quote your questionnaire with one click .

At the same time, there is a strategy for users to fill in the questionnaire ：

Hematemesis collection ！！ How to get more people to fill in your questionnaire ？

4.4 Questionnaire data processing and analysis

4.4.1 Questionnaire data analysis process

step1： User profile analysis ： The sample is representative

After collecting the questionnaire , We need to verify that we understand the user profile of the sample group , Clear the characteristic attributes of the survey sample group , Accurately position and investigate the customer base , Can be analyzed by contingency table 、 Frequency analysis 、 Descriptive statistics, etc

Contingency table analysis

By analyzing sample portraits , We can filter out better ones by setting some conditions 、 More representative groups , Improve the reliability and validity of the questionnaire .

For example, a cognitive survey data about the meta universe done before , Collected 505 Questionnaire data , However, I found that the reliability of the questionnaire was very low , in other words , The survey group of this questionnaire has a wide variety of opinions , There is no way to form a consensus , For example, as shown in the figure below ：

User profile analysis

The figure shows the proportion of men 50.89%, Female occupation 49.11%, There is little gender difference , In the age distribution 18~40 The proportion of years old is the highest , among 18 Under the age of (0.396%) The minimum , Most of the survey users are office workers , The share has reached 82.7%, At the same time, the monthly income is 1~2w The range reaches 39.41%, At the same time, college degree or above has reached 90.69%, Therefore, the majority of white-collar survey users .

The intra group correlation coefficient is the consistency of the evaluation of a group of objects in the questionnaire survey .

adopt SPSSPRO One key analysis of the intra group correlation coefficient , Answer the questions in the questionnaire , The relevance of the survey sample results is only 0.224, That means that the views of non white-collar groups and white-collar groups are different , Therefore, it is necessary to eliminate non white-collar groups .

So by SPSSPRO Filter settings for , Set the filter criteria as ：

Age is 18~40 year
As an office worker
The highest education background is junior college or above
Monthly income 1W above
IP The location is located in the five first tier cities of Beijing, Shanghai, Guangzhou, Shenzhen and Hangzhou

Data remaining after filtering 187 Samples , After the intra group correlation coefficient was repeated, the correlation degree reached 0.916, At the same time, through reliability analysis ,Cronbach’s α The coefficient is 0.951, It shows that the reliability of the questionnaire of white-collar group is excellent , The consistency of investigation and evaluation is high .

In reality, the questionnaire survey groups we collected are certainly diverse , Therefore, the survey group user portrait analysis of the questionnaire , It is the most important to conduct questionnaire analysis 、 It is also the most critical first step .

step2： Reliability and validity test ： The survey tool is stable

Reliability analysis ： Here's the problem ： Analyze whether the questionnaire is credible ？ Whether the consensus of people is unified ？

Example ：

Measure the existing one collected by 12 Customer satisfaction scale , Whether the measurement result is reliable .

Cronbach Reliability analysis is the most common , One of the most widely used measurement methods , Use one indicator directly, that is Cronbach The value of the reliability coefficient is used to describe the reliability level . if Cronbach The value of the reliability coefficient is greater than 0.6, Generally speaking, the reliability is acceptable , The larger the value of the reliability coefficient, the better .

except Cronbach reliability , Another kind of reliability is called half reliability , The principle is to analyze the items ‘ Split ’ In two parts , Then look at the half factor value , If the value of the half factor is greater than 0.6 The above description is acceptable , The bigger the better .

besides , Correlation analysis can also be used to measure reliability , For example, test-retest reliability can be measured by correlation analysis , First measure the data , Measure the data again at intervals , Carry out correlation analysis on the two data , The higher the correlation coefficient , The test-retest reliability is better .

If it is an experimental study or evaluator data , Usually in medical research , You may also use ICC Intra group correlation coefficient , The purpose is to study the similarity of data , Or consistency , Sometimes the measurements are repeated , Or the evaluation and scoring data will also be used ICC The intra group correlation coefficient was used to measure the reliability , if ICC Greater than 0.6, Generally, the data consistency is acceptable ,ICC The higher the value, the better .

If the data is not suitable for statistical analysis , That is, use the analytical method , Use an indicator to measure the level of reliability . Then describe it in words , It is also possible to prove that the data is reliable . For example, the data has been processed for outliers , Set invalid samples for data , The rest of the data is reliable and true . besides , Describe in detail the data collection and processing process , It is also a way to prove the reliability of data .

To make a long story short , As long as we can prove that the data is reliable and true （ Including analytical methods , Description, etc ）, It can be used to demonstrate the reliability level of data .

Validity Analysis ： Whether the design topic is reasonable and effective ？

Example ：

Measure the existing one collected by 12 Customer satisfaction scale , Measure whether the design of the topic is consistent with

The reason is . The following figure , Scale questions “ Customer satisfaction （ air ）” In the overall in factor 2 On , Other scale questions are all on the same factor , The title of the scale is （ Customer satisfaction （ air ）） Unsatisfied validity , It is suggested to eliminate .

Validity can generally be divided into content validity 、 Criterion validity 、 Structural validity . among , Content validity refers to whether the content of the questionnaire design can meet the hypothesis and purpose of the study ; Criterion validity refers to the , We usually set multiple topics for one research problem , Criterion validity refers to whether these questions are consistent with the relevance of the validity standard questions we set , Generally, Pearson correlation analysis can be used , As shown below ：

Criteria of judgment ： The greater the correlation coefficient , It means the higher the correlation , The higher the validity .

Structural validity refers to the load coefficient in factor analysis , Whether the measurement items of the same type are on the same factor , Exploratory factor analysis is commonly used 、 Confirmatory factor analysis .

SPSSPRO Two methods provided

① Exploratory factor analysis

Operation method ： Use SPSSPRO[ Validity Analysis ] or [ Factor analysis （ Exploratory ）] Function analysis , Put in the scale ;

② Confirmatory factor analysis

Operation method ： All items under the same factor should be put into the same analysis box , And so on , Several dimensions are put into several analysis boxes .

step3： Descriptive statistics ： Sample characteristic behavior analysis

The validity of the sample group has been verified 、 The reliability and validity of the questionnaire , The next step is to analyze the response of the research group to each option 、 The selection of the scale .

Generally speaking , Some hypothesis tests and various cross analyses are mainly used here , Among them, the core is multiple response cross analysis and multiple response frequency analysis , It's a little convoluted , To put it simply ：

Multiple response frequency analysis ： I study multiple choice questions , The selection proportion of the survey group .

Multiple response cross analysis ： It studies the cross selection of various topics .

Multiple response frequency analysis 【 Multiple choice analysis 】

In fact, the first multi topic is the grouping item , Study the selection of the second multi topic .

Multiple response frequencies （ cross ） analysis 【 Multiple choice questions are grouped items , The single choice is the analysis item 】

Multiple response frequencies （ cross ） analysis 【 Single choice questions are grouped items , Multiple choice questions are analysis items 】

step4： Inferential statistics ： Research core

Inferential statistics is the core of questionnaire analysis , Because only here can we really use some more in-depth methods , For example, regression analysis , Hypothesis testing and comprehensive evaluation . According to the type of statistical model, it can be divided into the following 6 Class algorithm ：

Difference test

A single sample means ：

Compare the difference between the sample data and a specific value

Such as ： A food production enterprise mainly produces canned food , The standard weight of each bottle should be 100g. To monitor the canning process , The quality inspection department of the enterprise shall conduct sampling inspection , To analyze whether the weight of each bottle meets the requirements . Now we randomly selected from a batch of food produced on a certain day 50 bottle , Measured weight of each bottle . Test whether the weight of this batch of food meets the requirements

Paired samples refer to ：

Whether the results of the same batch of samples at different times are different , Mathematically speaking, it is a continuous variable with the same number of samples X1 And continuous variables X2 The difference between .

Such as ： Test a hospital 30 Whether the blood pressure of patients is consistent before and after injection of a certain drug

Independent samples refer to ：

Analyze whether there is obvious difference between a definite variable and one or more quantitative variables , Mathematically, it means that there are multiple groups of continuous variables with different sample numbers X1、X2、X3...... The difference between .

Such as : Study students from different schools ( The number of students in each school is not necessarily equal ) Whether there are differences in the scores of the college entrance examination .

Post hoc multiple analysis refers to ：

Know multiple sets of continuous variables X1、X2、X3...... There are differences between , Want to explore which two variables are different ？

Such as : Students from different schools are known ( The number of students in each school is not necessarily equal ) There are differences in college entrance examination scores , Explore which two schools are different .

Correlation analysis

Correlation analysis is to analyze the degree of correlation between two variables . There are three ways to calculate correlation analysis , Namely Pearson The correlation coefficient （ Applicable to quantitative data , And the data satisfy the normal distribution ）、Spearman The correlation coefficient （ Use when the data does not satisfy the normal distribution ）.Kendall's tau -b The correlation coefficient （ Ordered categorical variables ）.

Example ：

Between height and weight ; The correlation between the relative humidity in the air and the rainfall is the problem of correlation analysis

Input / output description

Input ： Two or more quantitative variables or ordered categorical variables

Output ： Whether there is significant similarity between the two variables and the degree of similarity

Aggregation of indicators

Index aggregation refers to the combination of multiple choice variables into the same type of factors .

After aggregation, indicators are usually used for linear regression or mediation 、 Adjust the 、 Path analysis .

regression analysis

Regression analysis includes linear regression in statistics , Tree model regression in machine school , Such as xgboost、 Decision tree and random forest, etc , And the return of deep schools , It's regression analysis in mathematical statistics , A statistical analysis method to determine the quantitative relationship between two or more variables . Such as in linear regression analysis , Include only one independent variable and one dependent variable , And the relationship between them can be approximately expressed by a straight line , This kind of regression analysis is called univariate linear regression analysis . If the regression analysis includes two or more independent variables , And the relationship between dependent variable and independent variable is linear , It is called multiple linear regression analysis .

Example ：

Through independent variables （ House age 、 Is there an elevator 、 Floor height 、 Room square ） Fitting the predicted dependent variable （ housing price ）

Input / output description

Input ： The independent variables X At least one or more quantitative or categorical variables , The dependent variable Y Quantitative variables are required （ If it is a variable of fixed class , Please use logistic regression ）.

Output ： The result of model test goodness , Linear relationship between independent variable and dependent variable, etc

Mediation

The study of mediation refers to the study of independent variables X Influence dependent variable Y when , Whether it will affect the relationship through intermediary variables .

The study of regulation refers to the study of independent variables X Influence dependent variable Y when , Whether different regulatory variables have different effects on the above relationship .

Such models are usually based on the aggregation of indicators , Analyze . For example, combine multiple-choice questions into working conditions , Analyze company satisfaction, etc

4.4.2 Questionnaire data analysis framework

Affect the research framework of relationship

Analysis steps

First, analyze the collected data by user portrait , Through frequency analysis of basic algorithm 、 Descriptive analysis , Statistical gender , Age , What is the distribution of educational background ; Through the basic algorithm contingency table analysis （ Chi square analysis ） Or subtotals , Analyze the age distribution of different degrees , Age distribution of different genders, etc , If the research group cannot be accurately identified , This means that the survey results can not form a unified understanding of the group , This means that the reliability of the questionnaire is extremely low .
After determining the user portrait , It can be verified by reliability analysis , Reliability indicators are mostly expressed by correlation coefficient , It mainly inspects the reliability of the questionnaire , It is a foundation of data analysis , Generally speaking , The reliability analysis of the questionnaire is only for the scale , Gender, age and other background information are quite different , Therefore, it is generally not included in the reliability analysis .
When the reliability is OK , What needs to be analyzed next is the validity of the questionnaire , That is, whether the questionnaire design is reasonable or not , Generally speaking , We are based on research purposes , Multiple topics will be set to collect opinions , Statistically speaking , The linear correlation of these problems will be very high , So after factor analysis , Will be incorporated into a factor component （ research objective ） Upper , If a multiple-choice question is not on this factor component at this time , There is a problem with the design of this topic ;
After validity analysis , Because we are based on research purposes , Multiple topics will be set to collect opinions , Multiple topics are actually linearly related , They all express the same view , Therefore, these topics can be condensed into a whole through index classification and analysis , At this time, factor analysis or principal component analysis can be used for factor classification and concentration , For later regression analysis .
Condense the redundant topics , At this time, we do not need to analyze multiple topics with the same opinions , Therefore, these concentrated factors can be carried out （ research objective ） The analysis of , This step is to add sample background to these data , For example, the views of different genders on life satisfaction in the city, etc ;
The above is the descriptive statistics of all data , It shows the opinions of the questionnaire data , The influence questionnaire survey usually analyzes the influence relationship through linear regression , But before , We need to verify whether there is a linear correlation between the data included in the independent variables , With relevant premises , It makes sense to study regression again ;
Linear regression is a fitting of a series of influencing factors and results , Fit an equation （ There is no equation for nonlinear regression , The human brain is like a nonlinear regression ）, Then by applying this equation to other similar events , You can predict , The so-called return , It is the trend towards a certain ideal state or equilibrium state , Through regression, we can find out which influencing factors , The law of influence on the results ;
Last , We also need to compare the views of different groups as much as possible , This requires a user portrait based on the first step , In reality , We collected a wide variety of questionnaire survey groups , We can analyze the views of different people on the impact results as much as possible by setting conditional screening , Analyze whether there are differences , Verify that your results are universal and so on .

Policy status research framework

The core is the user portrait “ grouping ” Research , Explore the cognition of different groups

Analysis steps

First, analyze the collected data by user portrait , Through frequency analysis of basic algorithm 、 Descriptive analysis , Statistical gender , Age , What is the distribution of educational background ; Through the basic algorithm contingency table analysis （ Chi square analysis ） Or subtotals , Analyze the age distribution of different degrees , Age distribution of different genders, etc , If the research group cannot be accurately identified , This means that the survey results can not form a unified understanding of the group , This means that the reliability of the questionnaire is extremely low .
After determining the user portrait , It can be verified by reliability analysis , Reliability indicators are mostly expressed by correlation coefficient , It mainly inspects the reliability of the questionnaire , It is a foundation of data analysis , Generally speaking , The reliability analysis of the questionnaire is only for the scale , Gender, age and other background information are quite different , Therefore, it is generally not included in the reliability analysis .
When the reliability is OK , What needs to be analyzed next is the validity of the questionnaire , That is, whether the questionnaire design is reasonable or not , Generally speaking , We are based on research purposes , Multiple topics will be set to collect opinions , Statistically speaking , The linear correlation of these problems will be very high , So after factor analysis , Will be incorporated into a factor component （ research objective ） Upper , If a multiple-choice question is not on this factor component at this time , There is a problem with the design of this topic ;
After validity analysis , Because we are based on research purposes , Multiple topics will be set to collect opinions , Multiple topics are actually linearly related , They all express the same view , Therefore, these topics can be condensed into a whole through index classification and analysis , At this time, factor analysis or principal component analysis can be used for factor classification and concentration , Condense the redundant topics , At this time, we do not need to analyze multiple topics with the same opinions , Therefore, these concentrated factors can be carried out （ research objective ） The analysis of ;
The overall sample analysis is to analyze the overall cognition of the questionnaire , This step is to add sample background to these data , For example, the views of different genders on life satisfaction in the city, etc ;
Grouping sample analysis is done by setting different background conditions , For example, men 、 Bachelor degree or above 、24 year ~40 year , Carry out customer group analysis , Explore the cognition of different groups ;
The difference test needs to use scientific methods to verify whether the views of different groups are different , This requires a user portrait based on the first step , In reality , We collected a wide variety of questionnaire survey groups , We can analyze the views of different people on the impact results as much as possible by setting conditional screening , Analyze whether there are differences , Verify that your results are universal and so on .
If we want to study the influence relationship , For example, research “ cognition ”,“ attitude ” Respectively for “ Behavior ” Influence relationship of , In this case, regression analysis can be considered , Linear regression is a fitting of a series of influencing factors and results , Fit an equation （ There is no equation for nonlinear regression , The human brain is like a nonlinear regression ）, Then by applying this equation to other similar events , You can predict , The so-called return , It is the trend towards a certain ideal state or equilibrium state , Through regression, we can find out which influencing factors , The law of influence on the results .

Adjust the / Intermediary research framework

Analysis steps ：

First, analyze the collected data by user portrait , Through frequency analysis of basic algorithm 、 Descriptive analysis , Statistical gender , Age , What is the distribution of educational background ; Through the basic algorithm contingency table analysis （ Chi square analysis ） Or subtotals , Analyze the age distribution of different degrees , Age distribution of different genders, etc , If the research group cannot be accurately identified , This means that the survey results can not form a unified understanding of the group , This means that the reliability of the questionnaire is extremely low .
After determining the user portrait , It can be verified by reliability analysis , Reliability indicators are mostly expressed by correlation coefficient , It mainly inspects the reliability of the questionnaire , It is a foundation of data analysis , Generally speaking , The reliability analysis of the questionnaire is only for the scale , Gender, age and other background information are quite different , Therefore, it is generally not included in the reliability analysis .
When the reliability is OK , What needs to be analyzed next is the validity of the questionnaire , That is, whether the questionnaire design is reasonable or not , Generally speaking , We are based on research purposes , Multiple topics will be set to collect opinions , Statistically speaking , The linear correlation of these problems will be very high , So after factor analysis , Will be incorporated into a factor component （ research objective ） Upper , If a multiple-choice question is not on this factor component at this time , There is a problem with the design of this topic ;
After validity analysis , Because we are based on research purposes , Multiple topics will be set to collect opinions , Multiple topics are actually linearly related , They all express the same view , Therefore, these topics can be condensed into a whole through index classification and analysis , At this time, factor analysis or principal component analysis can be used for factor classification and concentration , For later regression analysis .
Condense the redundant topics , At this time, we do not need to analyze multiple topics with the same opinions , Therefore, these concentrated factors can be carried out （ research objective ） The analysis of , This step is to add sample background to these data , For example, the views of different genders on life satisfaction in the city, etc ;
The above is the descriptive statistics of all data , It shows the opinions of the questionnaire data , The mediation 、 The questionnaire survey of adjustment model usually analyzes the relationship through regression analysis , So before that , We also need to verify whether there is a linear correlation between the data included in the independent variables , With relevant premises , Then study the intermediary 、 Adjustment makes sense ;
Mediation is the study of X Yes Y In the process of , Whether to mediate variables M It acts as an intermediary bridge , namely X->M->Y Such a relational structure . And the regulatory role is to study X Yes Y In the process of , Different regulatory variables Z when ,X Yes Y Whether there is a significant difference in the degree of influence . Both mediation and regulation are used for hypothesis testing , The two have something in common , There are also substantial differences . The study of mediation and regulation needs to be carried out separately , In fact, the underlying mediation models are all based on linear regression , Linear regression is a fitting of a series of influencing factors and results , Fit an equation （ There is no equation for nonlinear regression , The human brain is like a nonlinear regression ）, Then by applying this equation to other similar events , You can predict , The so-called return , It is the trend towards a certain ideal state or equilibrium state , Through regression, we can find out which influencing factors , The law of influence on the results ;
Last , We also need to compare the views of different groups as much as possible , This requires a user portrait based on the first step , In reality , We collected a wide variety of questionnaire survey groups , We can analyze the views of different people on the results as much as possible by setting conditional screening , Analyze whether there are differences , Verify that your results are universal and so on .

Research framework of experimental differences

Experimental difference research refers to research methods with similar experimental research background , For example, study whether new drugs are effective , There are generally two ways ：

Experimental control group （ Independent samples ）, The experimental mice were divided into two groups , One group is the experimental group , The other group was the control group , The experimental group was injected with new drugs , The control group was not treated .
Before and after the experiment （ sample ）, For example, measure the injection of new drugs in experimental mice ( experiment ) front , Compared with the blood pressure after the injection of new drugs .

Analysis steps

First, analyze the collected data by user portrait , Through frequency analysis of basic algorithm 、 Descriptive analysis , Statistical gender , Age , What is the distribution of educational background ; Through the basic algorithm contingency table analysis （ Chi square analysis ） Or subtotals , Analyze the age distribution of different degrees , Age distribution of different genders, etc , If the research group cannot be accurately identified , This means that the survey results can not form a unified understanding of the group , This means that the reliability of the questionnaire is extremely low .
After determining the user portrait , It can be verified by reliability analysis , Reliability indicators are mostly expressed by correlation coefficient , It mainly inspects the reliability of the questionnaire , It is a foundation of data analysis , Generally speaking , The reliability analysis of the questionnaire is only for the scale , Gender, age and other background information are quite different , Therefore, it is generally not included in the reliability analysis .
When the reliability is OK , What needs to be analyzed next is the validity of the questionnaire , That is, whether the questionnaire design is reasonable or not , Generally speaking , We are based on research purposes , Multiple topics will be set to collect opinions , Statistically speaking , The linear correlation of these problems will be very high , So after factor analysis , Will be incorporated into a factor component （ research objective ） Upper , If a multiple-choice question is not on this factor component at this time , There is a problem with the design of this topic ;
After validity analysis , Because we are based on research purposes , Multiple topics will be set to collect opinions , Multiple topics are actually linearly related , They all express the same view , Therefore, these topics can be condensed into a whole through index classification and analysis , At this time, factor analysis or principal component analysis can be used for factor classification and concentration , For later regression analysis .
Condense the redundant topics , At this time, we do not need to analyze multiple topics with the same opinions , Therefore, these concentrated factors can be carried out （ research objective ） The analysis of , This step is to add sample background to these data , For example, the views of different genders on life satisfaction in the city, etc ;
The above is the descriptive statistics of all data , It shows the opinions of the questionnaire data , Experimental questionnaire survey is usually conducted through difference analysis , Difference analysis is a common data analysis method , A method for detecting whether there is a difference between the experimental group and the control group and whether the difference is significant in scientific experiments . also called 【 Significant difference test 】, It is a kind of hypothesis test ;
It should be noted that , Experimental analysis sometimes needs to pay attention to the analysis of interaction , For example, the same land ,A Fertilizer increases production 40KG,B Fertilizer increases production 50KG,AB Increase production with chemical fertilizers 100KG, This is the extra 10KG It's interaction ;
In interaction analysis , The first thing to do is to test the arguments X Whether it is significant , For example, does the above example include A Chemical fertilizer and whether it is added B Whether the two variables of chemical fertilizer are significant , If the independent variable shows significance , It means adding chemical fertilizer A、B Impact on production . Interaction study refers to the study of two categorical variables for dependent variables Y( Continuous variable ) Influence , Two categorical variables X At different levels , Categorical arguments X about Y Differences in the extent of impact of . For example, research into A Fertilizer and addition B The effect of chemical fertilizer on production . This is the time to consider , In the absence of joining A Fertilizers and additives A Two cases of fertilizer , Research adds B Whether the effect of chemical fertilizer on production is different . And whether to add B Two cases of fertilizer , Research adds A Whether the effect of chemical fertilizer on production is different , If neither of the two categorical independent variables is significant , The interaction study ends , Because neither of the two independent variables is related to the dependent variable Y An impact , It is even less likely that there will be interaction studies . If two categorical arguments X There is one of them. , Or both are significant , It is possible to conduct subsequent interaction studies . When two categorical arguments X There is one of them. , Or both are significant , Then we should analyze the significance of the interaction items , If the interaction item is significant , It means there is interaction , If there are interactions, you can continue to use the interaction graph （ Parallel or not ） Conduct visual analysis .
Last , After the above interaction study , If the independent variable X Showing significance , Researchers can continue to compare the independent variables in depth X At different levels, the dependent variable Y The details of , Further mining details .

4.5 Write reports and PPT

Report outline

Report format

The format of the report shall be in strict accordance with the requirements of the competition documents , The chart format is reasonable , Be able to use three line table and chart table ;

Master reporting requirements ：

1） Title page

This item must have . It includes the name of the project , However, the name of the school shall not appear 、 Team name 、 Names of team members and instructors .

2） Catalog

This item must have . Help judges and readers find the information they need .

3） List of tables and illustrations

This item must have . Present the results of the survey in a visual form such as a chart , At the same time, it is easy to compare with data from other sources , Give the catalog for the convenience of readers .

4） A point or summary

This item should have . The summary usually does not exceed 2 To 3 page , It reveals the most important findings and main conclusions of the survey . For those who don't have time to study all the details of the main report and want to know the main points of the findings , This is a simple and important reference material . It very briefly describes the objectives of the survey 、 Time of investigation 、 The location of the investigation and the topics covered by the investigation . Then describe the most valuable findings paragraph by paragraph in a short paragraph , Start with the overall results , To some more specific 、 Unexpected results that need to be emphasized . Of course , The expression of this part is segmented text or simple （ Table of contents ） summary , There is no uniform provision .

5) introduction

There should be an introduction at the beginning of the main body of the general report . The introduction gives the background information of the investigation , Such as the source of the project 、 Sponsor unit 、 Research objectives 、 Brief description of the method and the purpose of this report , Maybe you need to give a project outline , Explain what has been done , And what has been studied , But these only need a general description , Because the detailed results will be given in the following chapters . The introduction also needs to make a brief introduction to the contents of the following chapters and the links between them .

6） The subject of the report

The main body of the report needs to be divided into several chapters . The contents that should be included in the main body of the report mainly include , Concepts related to this survey and definitions of important indicators ; A description of the methods used in the survey ; Summary of the basic information data of the respondents ; Description of survey data quality ; Findings, etc . Each chapter topic should be highlighted , The language should be as concise and clear as possible .

7） Conclusions and recommendations

The conclusion is necessary , Advice is not necessary . Special attention should be paid to the survey data and analysis of data from the team for conclusions and recommendations , Not from other literature or references .

8） Bibliography

All references should be clearly marked .

9） appendix

The appendix shall describe the investigation topics not covered by the report subject and the investigation methods used 、 statistical software 、 Data processing methods and other supplementary descriptions not covered in the main report , Lest the text be too long . However, it should be noted that the appendix in this competition report should not be too long , Because the number of words in the appendix is also included in the total number of words .

The number of words and typesetting format of the report ：

1． The full text of the competition report is 1-1.5 swastika .

2． Except for the title page , Don't directly show the information of the school and the team on other pages .

3． All articles adopt WORD file ,A4 Paper layout .

Title and body ： The general title of the report body is bold in Song typeface , Single row spacing , Anterior segment 、 Select automatic spacing after segment ; The first level title in the report is in bold small three characters ; The secondary title is in bold four point characters ; Other titles and text are in Song typeface small four point characters . Abstract 、 References and other names are in bold font , The content is Song typeface small four character . Row spacing 1.25 times , Pre segment spacing 0.5 That's ok . The title of the chart in the text is in Song typeface , The words in the table are generally Song typeface small four or five characters , Single row spacing . Avoid using variant words 、 Compound words and all non-standard simplified words , Unless necessary , Do not use traditional characters .
Page setup and layout requirements ： All articles are written according to A4 Paper layout . The margins are WORD System default margin , namely ： The upper and lower margins are 2.54 centimeter , The left and right margins are 3.18 centimeter .
The main content of the report should be concise 、 clear , Not too many levels , The level serial number is ： One 、 ;（ One ） ; 1. ; （1）; 1）.
Number usage ： The Gregorian calendar century 、 years 、 year 、 month 、 Japan 、 moment 、 All kinds of counting 、 All measurements are in Arabic numerals ; Chinese characters were used in the summer calendar and the historical chronicles before the Qing Dynasty , And the year of A.D. is marked in parentheses ; An approximate number represented by two adjacent numbers in parallel , Use Chinese characters .
Figures in the text, etc ： The charts in the text 、 appendix 、 reference 、 All formulas are numbered consecutively in Arabic numerals . Pictured 1, surface 1, Note appended 1, The formula （1）. The sequence and title of the graph are placed at the bottom of the graph and centered , The table sequence and table questions are placed at the top of the table in the middle , Between diagram sequence and diagram question 、 There are two blank spaces between the table order and the table question . The formula number in the paper is enclosed in parentheses and written at the end of the right line , There is no dotted line .
reference ： To the author of the citation 、 Source 、 Notes on version and other details . Formats and examples ：

① Monograph format ： Main responsible person ． title [ Document type identification ]． Place of publication ： publisher , Year of publication ． Example ：[1] Chen Chaoyang , Wang Kezhong ． Organizational behavior [M]． Shanghai ： Shanghai University of Finance and Economics Press ,2001．

② Format of thesis collection ： author ． title [ Document type identification ]． The editor ． Collection name ． Place of publication ： publisher , Year of publication ． Example ：〔2〕 Liushousheng ． A study of the stages of Chinese history [A]． Guan Hong , Wei Ping ． The beginning of life problems —— Si Nian's academic essays [C]． Beijing ： China development press ,2001．

③ Journal article format ： Main responsible person ． title [ Document type identification ]． Title , year , volume （ period ）． Example ：[3] Wuwenyan ． Thoughts on the democratic management of employees and the establishment of harmonious enterprises [J]． Modern corporate culture , 2008,（17）．

④ Newspaper article format ： Main responsible person ． title [ Document type identification ]． The name of the newspaper , Publication date （ Edition ）． Example ：[4] Dingshixiu ． Construction management [N]． Construction daily ,2007-08-24 （08）．

⑤ Quote the article format of Internet websites ： author ． Article title ． website ． Release time .

Example ：[5] Wangjinying . A comparison of the role of human capital investment in economic growth between China and India ．http://www.amteam.org/ docs/bpwebsite.asp.2005 year 8 month 9 Promulgated by the .

◆ The same monograph 、 proceedings 、 Periodical 、 Newspaper articles , All use only one serial number , And the page number should be uniformly marked after the corresponding serial number in the article .

◆ Different types of references are marked with different capital letters , Such as monograph ：[M]; proceedings ：[C]; Newspaper articles ：[N]; Journal articles ：[J]; Dissertation ： [D]; The report ：[R]; standard ：[S]; Special issue ：[P].

notes ： Used to explain or explain a particular content in the text , Its serial number is ：①②③……, Comments are placed in the footer .

PPT Make

Take the report content as the template , Make one 10 Minute or so PPT, Pay attention here PPT At the roadshow , Be as creative as possible , Before the road show , A team made a before the presentation 30s A melodrama , It gave the judges and us a lot of attention （jingxia）, Not surprisingly, I got a high score , And the judges of the national competition also said that we should pay attention to teamwork , Not everyone here wants to talk about it , It is better for everyone to participate in the roadshow , The best way to solve this problem is to have only one speaker , Other people match a sitcom or a narration 、 Opening remarks .

online PPT There are a lot of tutorials , There is not much to be said here .

After finishing the first draft, we brainstorm together , Discuss and modify 、 Then invite the instructor to review , Put forward modification opinions . Finally, the team members modify it again .

, The answer of the national competition is that the judges and teachers will ask questions together , All the questions are over , The contestant will answer again . There may be several problems . It is recommended to bring a small notebook , Make a brief note of , Easy to answer and clear your mind .

4.6 Competition defense skills

The provincial contest defense is generally 10 Minute presentation ,5 Minute Q & A , A specially assigned person will be arranged next to the time , Time will interrupt , The national defense judges and teachers will ask questions together .

Uniform clothing , Remember ;
Don't memorize , Road show with emotion ;
A lecturer will perform the roadshow from the beginning , Don't divide into many people ;
Take part in , Except for the roadshow people , Others can design a sitcom at the beginning or end or intersperse it in the middle , Improve the roadshow performance ;
Pay attention to the unified answer format when answering questions , for example ： Thank you for your question , For your question , I would like to make the following answer ：…, That's my personal opinion , If there is any improper or wrong expression , Please criticize and correct me , Thank you, teacher ;

4.7 Share a questionnaire data analysis artifact

At present, the domestic data analysis product market mainly consists of SPSS、SAS、Matlab Wait for several old brand products to occupy , The threshold for downloading and using these software is high , Genuine software is expensive , It also needs to reserve some theoretical knowledge of mathematical statistics , Some software needs programming to operate , For those who do not have relevant basic knowledge , Long learning cycle .

Statistical software used to be used in school , The operation interface is in English , We also need to consult various literature to understand the steps of statistical analysis . Answer: the Lord always hopes to have a domestic product 、 Easy to operate data analysis products .

Under all kinds of coincidence , Answer: the Lord finally has a chance to realize his dream , Answer: domestic products mainly researched and developed 、 The free data analysis platform is called SPSSPRO, Use online SAAS Pattern , The platform makes the algorithm model black box , Users only need data input and output , The analysis report can be automatically generated , It simplifies the learning and operation process of data analysis .

“ such as , If using other software t test , Users first need to know what methods are available for the difference test , You need to learn how to operate the software step by step , Finally, understand the specific meaning of each parameter in the data results .SPSSPRO The application cases of each algorithm are listed , Through simple operation, the user , Drag the data to the corresponding analysis box , You can generate analysis results with one click , No need to learn the principles of analysis . meanwhile , For the results of the analysis , The platform also provides intelligent interpretation , Prompt the meaning of each parameter , Whether the model has passed the inspection, etc . Users no longer have to worry about not understanding the data coming out ”.

therefore , No need to learn complex mathematical formulas , You can master data analysis skills , Thus, data analysis is used to assist decision-making , Flexible application to specific scenarios .

Just like beautiful pictures , You don't need to understand PS The basic principle can be completed in a few seconds p chart . Answer: the Lord wants to pass SPSSPRO, People don't have to learn complicated mathematical formulas , You can master data analysis skills , Thus, data analysis is used to assist decision-making , Flexible application to specific scenarios , Lower the threshold of data analysis , Let everyone be a data analyst .

SPSSPRO- Free professional online data analysis platform

Strong professional

Covering all professional statistical algorithm models , In the process of early research and development , The team has accumulated 150 A variety of algorithm models , After the current product is reconstructed , Launched 30 A variety of common algorithms that are more close to users' operating habits and needs , It can meet the needs of college teachers and students 、 Demand for data cleaning and statistical analysis of market researchers and scientific researchers .

System one key generation analysis and interpretation , Automatic annotation of key terms , Can understand without professional knowledge .

Simple and practical

The operation procedure is simple , Just drag the data to the specified box , Click start analysis ; Seamless connection with mainstream online research platforms such as the questionnaire network , Easily realize professional online SPSS analysis

Open for free

spsspro Completely free , There is no need to pay , Use wechat to log in , All algorithm functions are free and open ; Don't download software , Do not use local computer resources , You can use it with a net .