Presentation is loading. Please wait.

Presentation is loading. Please wait.

The Basics of Social Science Research Methods

Similar presentations


Presentation on theme: "The Basics of Social Science Research Methods"— Presentation transcript:

1 The Basics of Social Science Research Methods
Susanne Barth, MSc

2 Structure of the lecture
Data collection and data analysis Self-study  Book Chapter 9, 10, 14 & 15 [Bha2012] Bhattacherjee, A. (2012). Social Science Research. Principles, Methods, and Practices. Tampa, Florida: University of South Florida. PDF

3 Your research process Research method: Validity/reliability & Pilot Testing Research execution: Data collection Data Analysis Case: Social Research Training Research report: lecture 23rd May 2017

4 Research method (1): Reliability/Validity & Pilot Testing (1)
Reliability: extent to which a measurement tool gives consistent results Validity: extent to which a measurement tool measures what it is supposed to measure Example indirect measurement: Volume as a measurement for weight  instrument to measure volume might give consistent results on volume = reliable  BUT is volume a valid indicator for weight? = Do a pre-test to figure out if volume is a valid measurement for weight Reliability: e.g. you weight yourself several times during a day and the scale does not vary much = high reliability you weight yourself several times during a day and the scale drastically varies = low reliability Validity: e.g. Intelligence test that actually measures intelligence = high validity Intelligence test that measures memory instead = low validity

5 Research method (2): Reliability/Validity & Pilot Testing (2)
Report about validity and reliability; what did you do to establish sufficient validity and reliability for your research: Have you used existing measurement scales? Have you performed a pre-study (pilot test) to test stimulus materials? Is the wording of the questions correct? Do you measure want you intent to measure? (validity) Gives your measurement consistent results? (internal consistency reliability)  Cronbach’s Alpha Calculation How did you select participants? Take into account possible threats regarding validity and reliability and address those threats in your methods and instruments section. Reliability: e.g. you weight yourself several times during a day and the scale does not vary much = high reliability you weight yourself several times during a day and the scale drastically varies = low reliability Validity: e.g. Intelligence test that actually measures intelligence = high validity Intelligence test that measures memory instead = low validity

6 Your research process Research method: validity and reliability
Research execution: Pilot Testing Data Collection Data Analysis Case: Social Research Training Research report: lecture 23rd May 2017

7 Data Collection (1) Type of data collection: qualitative vs. quantitative Expected amount of collected data Online questionnaire: Between 200 and 250 respondents Hard-copy questionnaire: Between 100 and 150 respondents Experimental designs: Depending on number of conditions (min 30 resp. per condition) Interviews: Between 15 and 20 participants Focus groups: Between 4 and 6 group sessions Content analysis: Depends on the research design Observations: Depends on the research design

8 Data Collection (2) Qualtrics (1)
Data collection procedure Online survey tool “Qualtrics” is available in cases of quantitative data collection (i.e. surveys, experiments, sorting tasks, heat maps...) Easy to use and distribute Direct export to SPSS

9 Data Collection (3) Qualtrics (2)

10 Data Collection (4) Qualtrics (3)

11 Data Collection (5) Qualtrics (4)

12 Data Collection (6) Qualtrics (5)
Many features and options: see video’s on YouTube for instruction and on Qualtrics’ website Information to register for an account can be found at:

13 Your research process Research method: validity and reliability
Research execution: Pilot Testing Data Collection Data Analysis Case: Social Research Training Research report: lecture 23rd May 2017

14 Make your data file

15 Data analysis (1) Make your data file (1)
Import your data to SPSS

16 Data analysis (2) Make your data file (2)
Clean your data! Make your data anonymously Only include the variables you need for your analysis in your final data file Translated student numbers into codes

17 Data analysis (3) Make your data file (3)
Data and variable view in SPSS

18 Data analysis (4) Make your data file (4)
Re-coding of variables Translated study programme into codes

19 Data analysis (5) Make your data file (5)
Re-coding of variables Re-coding of variable „smartphone ownership“

20 Data analysis (6) Make your data file (6)
SPSS descriminates between scale, nominal and ordinal measurement Scale: ordered categories, has a „true zero“ point e.g. age (if open question) (If you group by age e.g , etc. it would be a nominal variable)

21 Data analysis (7) Make your data file (7)
nominal: e.g. region, Zip-code, operating system, gender, yes/no (numbers that are simply used as identifiers)

22 Data analysis (8) Make your data file (8)
ordinal: ranking (e.g. Likert Scale)

23 Data analysis (9) Missing value (1)

24 Cronbach‘s alpha calculation

25 Data analysis (10) Cronbach’s Alpha (1)
Gives your measurement consistent results? (internal consistency reliability)  Cronbachs Alpha Calculation

26 Data analysis (11) Cronbach’s Alpha (2)

27 Data analysis (12) Cronbach’s Alpha (3)

28 Data analysis (13) Cronbach’s Alpha (4)
Cronbach‘s Alpha for the construct Perceived Security (PS) is .72 In the original scale .82 Look at „if item deleted“ Cronbach's alpha Internal consistency α ≥ 0.9 Excellent 0.9 > α ≥ 0.8 Good 0.8 > α ≥ 0.7 Acceptable 0.7 > α ≥ 0.6 Questionable 0.6 > α ≥ 0.5 Poor 0.5 > α Unacceptable

29 Data analysis (14) Cronbach’s Alpha (5)
Might consider to delete 4th item

30 Compute variable

31 Data analysis (15) Compute variable (1)

32 Data analysis (16) Compute variable (2)

33 Data analysis (17) Compute variable (3)

34 Data analysis (18) Compute variable (4)

35 Mean

36 Data analysis (19) Mean construct (1)

37 Data analysis (20) Mean construct (2)

38 Comparison of groups (ANOVA)

39 Data analysis (21) Comparison of groups (ANOVA) (1)
To what extent do females score differently on perceived security compared to males?

40 Data analysis (22) Comparison of groups (ANOVA) (2)

41 Data analysis (23) Comparison of groups (ANOVA) (3)
The difference between gender (female vs male) and the mean score on Perceived Security is not statistically significant (α = .642) We usually consider results as statistically significant at a threshold of .05 (or below)

42 Data analysis (24) Comparison of groups (ANOVA) (4)
Fundamentalists = are most protective and concerned about their privacy Pragmatists = weigh the potential pros and cons of sharing personal data, but are less concerned than fundamentalists Perceived Security of the app store the individual usually uses, measured on a Likert Scale going from 1 = completely disagree to 7 = completely agree Fundamentalist perceive the security of the app store they usually use as less secure than pragmatists (α < .05)

43 Descriptive statistics

44 Data analysis (25) Descriptive statistics (1)

45 Data analysis (26) Descriptive statistics (2)
Age: (SD 6.14) Average time s.o. has owned a smartphone (in months): (SD 36.15) Average time s.o. has owned a smartphone (in years): 6.66 (SD 3.01) etc. . Standard Deviation = amount of variation within a given data set A low standard deviation indicates that the data points tend to be close to the mean (also called the expected value) of the set, while a high standard deviation indicates that the data points are spread out over a wider range of values.

46 Data analysis (27) Frequencies (1)

47 Data analysis (28) Frequencies (2)
Gender: 82.9 % male and 17.1 % female Study program: 82.9 % Computer Science 8.6 % Electrical Engineering etc. Describe all the characteristics of your sample that you consider as important to mention! Be as specific as possible but do not report nonsense…

48 Data analysis And much more tests… Depends on your RQ‘s
Read Andy Fields‘ Book Have a look at Have a look at Go to methodology shop (methodologie winkel BMS) Ask the supervisor team if you need help [Fie2009] Field, A. (2009). Dicovering statistics using SPSS. LA: Sage.

49 Now: Self-study: e.g. developing/finalizing your questionnaire developing an analysis strategy working on your study design or experimental setting/manipulation etc… What you eventually will be working on depends on the status of your project.

50 © by WingedWolf


Download ppt "The Basics of Social Science Research Methods"

Similar presentations


Ads by Google