Summary: Data Analytics Ii

Read the summary and the most important questions on Data Analytics II

• 1 Lecture 1 - Powerpoints Quantitative Research

  This is a preview. There are 17 more flashcards available for chapter 1
  Show more cards here

• What is the importance of science?

  It makes statements about reality with a certain degree of certainty

• What is the role of statistics in quantitative research?

  It checks:
  
  

  • Whether your research supports your hypothesis or not;
  • How to value the relationships (egthe connections are strong or weak);
  • To discover (even more) patterns in your data (egFactor analysis);
  • To justify your method (egQuality of your scales, power);
  • Or to give unambiguous descriptions (IQ scores) 

• Main learnings abut the necessary insights on statistics

  
  

  1. Statistics is a means by which certain claims can be checked (in combination with research of course).
  2. Comparable groups are needed for a comparison between A & B (& C etc.) (Exam resit example).
  3. The greater the difference, the greater the likelihood that there really is a difference (numbers are not influenced by an accidental change, they have a meaning).
  4. The larger the groups are, the more confidence we have in this difference (decreases sensitivity to outliers). There is a difference between interviewing 10 people or 1000.

• What is the usefulness of knowledge about quantitative research?

  It helps to:
  

  • make better decisions (for the consumer / company)
  • better understand and evaluate research in the media/science

  • gives more general knowledge
  • better substantiated research - Project groups, research internship
  • become more critical through this knowledge

• What are the qualitative measurement levels?

  • Nominal: points to a characteristic without value distinction (man/woman)

  
  

  • Ordinal: arranges in an order with value (less/more, low/high, age categories)

• What are the quantitative measurement levels?

  Interval: represents an equal difference (temperature interval from 3 to 6 to 9 ºC). This makes differences meaningful and comparable. There is no absolute zero point, ie ratios are not meaningful
  
  Ratio: measurements start at a natural point of zero, intervals are similar and ratios are meaningful (length, age in years/months/days, weight, speed)

• What are the 2 types of hypotheses?

  
  

  1. Null-hypothesis (h0): there is no connection, no difference, no effect - "Money does not affect happiness"
  2. Alternative hypothesis (h1): there is a connection, difference, effect - "Money affects happiness positively"

• How to measure the correctness of an hypothesis

  • Assume the correctness of H0 - confirm through statistics whether H0 can be achieved (and therefore whether H1 can be rejected)
  • Conclude that H0 can be rejected (and whether H1 can be accepted). Instead of confirming H1, look for evidence that H0 is not valid (eg: my H1 is correct because H0 presents X errors/faults)

• What is statistical significance?

  A term that indicates in statistics whether a certain relationship or difference is coincidental or not.
  You test whether the difference you find in your sample is coincidental or whether the difference actually exists.

• When do we speak of statistical significance?

  The difference between our observation and our zero hypothesis is so great that you would not expect this difference to occur on the basis of chance (accidentally).