COMP3425 and COMP8410 Data Mining S1 2023 Assignment 2: Description of Data

Data and Metadata

The data supplied for the assignment arises from The Australian  Data Archive’s ANU  Poll Dataverse [1]. As a student of the course, you are assumed to accept the Terms and Conditions of Use reproduced below. Please read them carefully. The custodian of the data has requested you delete your data at the end of the course.

In particular the data captures the results of a survey poll conducted in 2019 on the topic of attitudes and behaviours towards Universities, amongst other things.  You can find a complete description of the purpose of the poll and coding of the data (metadata) and also a descriptive

summary of the poll results here:

https://dataverse.ada.edu.au/dataset.xhtml?persistentId=doi:10.26193/GOVGBB

The data is provided to you for the assignment in two forms. The first is the original dataset as downloaded from the ADA called 2.ANUPoll2019RoleOfGovernment_CSV_01445.csv, in comma-separated-values  format.   This    data    is    described    by    the    metadata    in    1. ADA.CODEBOOK.01445.xslx       and      the       corresponding       question      text       in       1. ADA.QUESTIONNAIRE.01445.pdf

The second is a form derived from the original, pre-processed for the COMP3425 data mining assignment, in comma-separated-values format called 3425_data.csv. Below you will find a description of the pre-processing undertaken and this, in addition to the original metadata, willbe needed to assist your understanding of the data.

If you are a COMP3425 (undergraduate) student, you must work with the pre-processed dataset 3425_data.csv.

If you are COMP8410 (postgraduate) student you may use either the original or the pre- processed data, or both. The original will give you more opportunity to show off your technical skills and creativity, while the  pre-processed one  is  more constrained  but  may save time, requiring you to spend less effort understanding the data, and helping to avoid some data errors. The same rubric will be used for marking in both cases, but the original dataset provides an extended learning experience and better opportunity for higher marks. Even if you use the original data, you may find it useful to observe the pre-processing that has been undertaken to produce 3425_data.csv  to seed ideas or to solve problems you encounter.

Pre-processing applied with Excel to derive 3425_data.csv

•   Only a selection of the original attributes have been retained.

•   The Q15_safe_gambler column has been added, based on respondent’s answers to    questions Q15a-i, which have answers that range from almost always to never.             Q15_safe_gambler is a normalized number in the range [0,1] that shows the rarity of the various problem gambling behaviours raised in Q15a-i. Refused and Don’t know   options are replaced by the midpoint value for each question,  and the field is null     when the Q15 questions were not asked.

Q15_safe_gambler = IF(NOT(Q14=" "),((IF(OR(Q15a=-98, Q15a =-99),2.5,    Q15a)+(IF(OR(Q15b=-98, Q15b =-99),2.5, Q15b))+(IF(OR(Q15c =-98, Q15c    =-99),2.5, Q15c))+(IF(OR(Q15d =-98, Q15d =-99),2.5, Q15d))+(IF(OR(Q15e    =-98, Q15e4=-99),2.5, Q15e))+(IF(OR(Q15f =-98, Q15f =-99),2.5,                     Q15f))+(IF(OR(Q15g =-98, Q15g=-99),2.5 Q15g))+(IF(OR(Q15h=-98, Q15h =- 99),2.5, Q15h))+(IF(OR(Q15i=-98, Q15i =-99),2.5, Q15i)))-9)/27,"")

•   The binary undecided voter column was added based on the given answer to Q4, and is TRUE when the answer to Q4 is one of -98, -99, 95, 97 and FALSE otherwise. That   is, IF(OR(OR(OR(Q4=-99, Q4=-98),Q4=95), Q4=97),TRUE,FALSE).

•    For two categorical columns, nominal Q2 and nominal StateMap, double quotation   marks were added to all non-empty cells. For the rest of the categorical columns,      you can use the same approach to help Rattle recognise categorical data in a column if necessary. For example, for nominal StateMap, the formula CONCATENATE("""",    StateMap, """") is used. For nominal Q2, the formula CONCATENATE("""",  TEXT(Q2, "0"), """") is used.

References

[1] Biddle, Nicholas; and Reddy, Karuna, 2019, “ANU Poll 2019: Role of the University”,

doi/10.26193/GOVGBB

Terms and Conditions of Use

This data has been distributed exclusively for students of COMP3425 and COMP8410 S1 2023 only. Data must be destroyed at the end of the course but may be re-obtained by  request to the Australian Data Archive.

Furthermore, fromhttps://dataverse.ada.edu.au/dataset.xhtml?persistentId=doi:10.26193/GOVGBB, I acknowledge that:

1. Use of the material is restricted to use for analytical purposes and that this means that I can only use the material to produce information of an analytical nature.

Examples of such uses are: (a) the manipulation of data to produce means, correlations or other    descriptive summary measures; (b) the estimation of population characteristics from sample data; (c) the use of data as input to mathematical models and for other types of analyses (e.g. factor      analysis); and (d) to provide graphical and pictorial representation of characteristics of the              population or sub-sets of the population.

2. The material is not to be used for any non-analytical purposes, or for commercial or financial gain, without the express written permission of the Australian Data Archive.

Examples of non-analytical purposes are: (a) transmitting or allowing access to the data in part or whole to any other person / Department / Organisation not a party to this undertaking; and (b)    attempting to match unit record data in whole or in part with any other information for the          purposes of attempting to identify individuals.

3. Outputs (such as statistics, tables and graphs) obtained from analysis of these data may be further disseminated provided that I: