Đang chuẩn bị liên kết để tải về tài liệu:
Mastering skills for the toefl ibt part 10

Tham khảo tài liệu 'mastering skills for the toefl ibt part 10', ngoại ngữ, toefl - ielts - toeic phục vụ nhu cầu học tập, nghiên cứu và làm việc hiệu quả | Mastering-Books_5 2006.5.29 2 13 PM Page 783 Writing Chapter 1 _ Skill A Practice 1 01 Step 1 Correlation Studies determine RELATIONSHIP BETWEEN two variables - researcher doesn t MANIPULATE variables - researchers MEASURE RATE at which variables change naturally Relationship types - Y increases when X increases POSITIVE CORRELATION - Y decreases when X increases NEGATIVE CORRELATION - sometimes a CAUSAL RELATIONSHIP can be inferred Step 2 Main point Correlation does NOT IMPLY causation - cannot be certain because investigators don t MANIPULATE VARIABLES - also a THIRD VARIABLE may be affecting the correlation ex. Eating ice cream and drowning have a POSITIVE CORRELATION - but a third variable is HOT WEATHER - Correlations can SUGGEST causal relationships but more RESEARCH is needed to prove it ex. A positive correlation between smoking and CANCER led to further research that proved a CAUSAL RELATIONSHIP Step 3 Reading - Main Idea Correlation studies can determine a connection between two variables. - Supporting Idea If the rate of one event increases when the rate of another event increases they have a positive correlation. - Supporting Idea If the rate of one event decreases when the rate of another event increases they have a negative correlation. Lecture - Main Idea Correlations found from correlation studies do not necessarily mean a causal relationship exists. - Supporting Idea Other third variables may be affecting the relationship between the two variables in a correlation study. - Supporting Idea Correlations found from correlation studies can suggest the need for further study to discover if a causal relationship truly exists. Step 4 CORRELATION STUDIES are useful tools because they describe relationships between different PHENOMENA as they occur in the natural world. It is important though that researchers be careful not to make the common erroneous assumption that a CORRELATION IMPLIES CAUSATION. Correlations indicate when two .