SPSS Statistics Lab Report

3/10/16

1

OBSERVATIONAL STUDY Lab 2

Out Study

¨ Are there differences in the ways students carry their books on a college campus? ¤ Chose a spot on campus ¤ Observed up to 30 students with your co-observer ¤ For each person observed:

n Verified with partner that you’re observing same person n Did no verify that you’ve made the same categorization

n Marked person’s apparent sex n Marked book carrying style that MOST closely fits

3/10/16

2

Our Study

¨ Sex and book carrying style ¤ Type of Observation? ¤ Type of Survey design? ¤ Scale of Measurement?

n Categorized sex (m or f) n 1=female, 2=male

n Categorized book carrying style

¤ Type of reliability most relevant?

1 2 3 4 5 6

Hanaway and Burghardt (1976)

¨ Differences: Hanaway and Burghardt (1976)

Research Methods (2016)

Samples:

Variables:

Categories:

Inter-rater Reliability:

3/10/16

3

Our Study

¨ Inter-rater reliability

¨ Summarize ¤ Mean, median, mode? ¤ Frequencies (histogram/bar graphs)

¨ Association ¤ Are sex and carrying style associated?

n HA: Sex and Carrying Style are associated (are not independent)

n H0: Sex and Carrying Style are not associated (are independent)

For Lab 2

¨ Title Page ¤ Running head: A short version of a title ¤ Centered: full title

¤ Centered: your name ¤ Centered: research affiliation

3/10/16

4

For Lab 2

¨ Method ¤ Describe in detail how the study was conducted ¤ Based on Method section, reader should be able to

replicate the study ¤ Three most common subsections:

n Participants: describe sample in general terms, numbers of males and females, approximate age range, demographics, (ONLY what you can safely say about sample)

n Materials: describe worksheet (describe each of 6 book-carrying categories)

n Procedure: describe observation technique; type of survey (cross- sectional/successive independent samples/longitudinal; # of observers, inter-rater reliability (% agreement between observers)

For Lab 2

¨ Results ¤ Actual findings of the study

n Descriptive/frequencies n # of males/females n Carrying style

n Analyses n Chi-square test of association

¤ Include data presented in tables or graphs n BUT figures own page at end!

3/10/16

5

For Lab 2

¨ Results (a recipe) ¤ Write down the question you are trying to answer

n “Here we describe book carrying behavior among college students.”

¤ Write down the analysis you’ve done, including variables n “We tabulated frequencies of book carrying styles and males and

females”

¤ Write the result of the analysis n “The resulting frequencies are depicted in Figure 1” n Bar graph on last page

For Lab 2

¨ Results (a recipe) ¤ Write down the question you are trying to answer

n “Is there an association between sex and book carrying behavior?”

¤ Write down the analysis you’ve done, including variables n “We computed a χ2 test contrasting the six book carrying behaviors

among male and female college students.”

¤ Write the result of the analysis n Statistic: “The resulting value, χ2 (df, n = # participants) = chi

square value , p < p-value , was(was not) statistically significant.” n The results in plain-English: “Women tended to carry their books...“

n Order of these last two steps can vary, but both need to be there for your results section to make sense

3/10/16

6

For Lab 2

¨ Figure 1 ¤ Bar graph from SPSS on own page

n Still has header with short title and page number

¤ Should be in grayscale (don’t rely on color to differentiate bars)

¤ Labeled UNDER bar graph

Figure 1. A short, detailed description of what’s in the graph

Chi-Square Info

¨ Degrees of freedom ¤ DF = (r - 1) * (c - 1)

¤ where r is # of levels for one categorical variable, and c is # of levels for other categorical variable

¨ Expected frequencies ¤ Computed separately for each level of one categorical variable at each level of other

categorical variable.

¤ Er,c = (nr * nc) / n

¤ where Er,c is expected frequency count for level r of Variable A and level c of Variable B, nris total # of sample observations at level r of Variable A, nc is the total # of sample observations at level c of Variable B, and n is the total sample size

¨ Test statistic ¤ Χ2 = Σ [ (Or,c - Er,c)2 / Er,c ]

¤ where Or,c is the observed frequency count at level r of Variable A and level c of Variable B, and Er,c is expected frequency count at level r of Variable A and level c of Variable B